Measuring Physical Density: Implications on the Use of Different Measures
on Land Use Policy in Singapore
By
Claire S. Chan
Bachelor of Arts (Honours)
National University of Singapore, 1993
Submitted to the Department of Urban Studies and Planning in partial fulfillment of the
requirements for the degree of
Master in City Planning
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
June 1999
@ 1999 Claire S. Chan, All Rights Reserved
The autorheo/gn
ChMbutc pubcly a
ocu1ift co.pos of ts thesis
cuinontin whobo or in pcit
Author
Department of Urban Studies and Planning
May 20 1999
Certified by
Assistant Professor Eran Ben-Joseph
Department of Urban Studies and Planning
Thesis Supervisor
Accepted by
Associate Professor Paul Smoke
Department of Urban Studies and Planning
Chair, MCP Committee
MASSACHUSETTS INSTITUTE
OF TECHNOLOGY
JUL 1 9 1999
LIBRARIES
Measuring Physical Density: Implications on the Use of Different Measures on Land Use
Policy in Singapore
By
Claire S. Chan
Submitted to the Department of Urban Studies and Planning
on May 20, 1999
in partial fulfillment of the requirements for
the Degree of Master of City Planning
ABSTRACT
Land use planning policies in Singapore have been driven by the goal to optimize the use
of the scarce land resource. Density measures how intensively land has been put to use
and is thus a good indicator of how planning policies are performing in relation to this
goal of land use optimization. This study looks at how physical density has been
measured in Singapore and assesses the suitability of the methods of measurement. It
also evaluates current planning practice to determine how successful they have been in
achieving the goal to optimize land use. The Development Guide Plan for the Punggol
Planning Area was used as the case study.
Thesis Supervisor: Eran Ben-Joseph
Title: Assistant Professor of Landscape Architecture and Planning
Acknowledgements
I would like to thank my thesis supervisor, Eran Ben-Joseph, for his guidance, patience
and encouragement in the course of this exercise. I am especially grateful for his
sensitivity in trying to understand my background and appreciate my perspective of the
topic.
My sincere appreciation also goes to my thesis reader, John de Monchaux, who
challenged me to undertake this study and has provided invaluable guidance and
feedback. I have truly benefited from his experience and insights on this topic.
My thanks is due to URA colleagues, Heng Siok Ngo, Han Yong Hoe and Smita
Choudhuri, who have kindly shared their extensive knowledge on physical planning with
me and given their time to answer the questions that I raised. I also am grateful Susan
Ang, Kevin Wee and Tan Soo Li, who were always an email away when I needed help
with background information.
To the many prayer warriors who have remembered me in their prayers, and to my family
and friends who have encouraged me, I say thank you. I have been greatly blessed that
you have been a part of my life.
Most of all, I thank God for the fullness of His grace. For in Him, I have truly been
enriched in every way. To God be the glory!
Contents
Abstract
Acknowledgements
Contents
List of Figures
List of Tables
Preface
1
Introduction
1.1
Land Use Optimization and Density
1.2
Scope
1.3
Aim
1.4
Methodology
1.5
Overview of the Thesis
1.6
Limitations
2
Land Use Planning Policies and Physical Planning in Singapore
2.1
Development Goals and Planning Policies in Singapore
2.2
Performance Indicators
2.3
Physical Planning in Singapore
2.4
Planning Approaches
2.5
The Planning Process and the Plans
3
Measuring Physical Density
3.1
Measured Density
3.2
Measuring Density with Ratios
3.3
Generic Density Measures in Ratios
3.4
Surrogate Measures of Density
3.5
Density Measures in the Singapore Context
3.6
Planning Standards in Singapore
Analyzing Land Use Planning Policies
Case Study - The Development Guide Plan for Punggol
4.1
Planning Area
Planning Policies
4.2
Planning Policies: Evaluating Successes and Failures
4.3
71
5
Strategies to Optimize the Use of Land
Classification of Housing Variety
5.1
Land Consumption
5.2
101
101
105
6
Findings and Recommendations
Findings and Recommendations
6.1
Questions for Further Study
6.2
Shortcomings
6.3
Conclusion
6.4
125
125
132
133
134
4
Bibliography
71
75
78
137
List of Figures
Population and Population Density - Singapore and
World Cities
Growth of Land Area and Urbanized Area (1970 to Year
X)
Relationship between Different Types of Density
19
21
2-2
2-3
2-4
Relationship between Goals, Policies and Performance
Indicators
Singapore's Concept Plans
Singapore's Planning Approach
The Planning Process
27
36
38
39
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
The Land Base (Molar Measure)
Floor Area Ratio (FAR)
Building Coverage
Setbacks
Population and Urban Densities (1970 to Year X)
Singapore's Planning Areas and Regions
The Quantitative Approach
Width of Travel Lanes
46
51
52
52
54
55
56
67
4-1
Location of Punggol Planning Area and Pasir Ris
Planning Area
Study Area for Pasir Ris DGP
Floor Area Ratios - Punggol and Pasir Ris DGPs
Net Housing Yield - Proposed (Punggol and Pasir Ris
DGPs) and Control by Floor Area Ratio
Housing Variety by Land Tenure
Housing Variety by Design Quality (Public Housing)
Clustering of Community Facilities
Community Facilities - Punggol and Pasir Ris DGPs
Public Transportation Network - Punggol DGP
MRT and LRT Network and Density - Punggol DGP
Green Connector Along Sungei Api Api in Pasir Ris New
Town
1-1
1-2
1-3
2-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
4-10
4-11
18
73
75
80
83
84
85
87
89
90
92
93
4-12
5-1
5-2
5-3
5-4
5-5
5-6
5-7
5-8
Park and Open Space Network with Green Connectors Punggol and Pasir Ris DGPs
Housing Variety Matrix
Housing Variety by Location
Level of Urbanization (1970 to Year X)
a) Distribution of Housing Types
b) Housing Type by Land Area
Typical Layout of Public Housing in Punggol 21
Road Network - Punggol and Pasir Ris DGPs
Road Types - Punggol DGP
Typical Road Reserve (Section)
96
102
104
107
110
112
115
116
117
List of Tables
2-1
Development Goals and Planning Policies
32-33
3-1
3-2
Defining the Boundary of Land
Comparative Density Measures Reflecting Population and
Housing Units
Population and Urban Densities (1970 to Year X)
Land Use Distribution at the City-Scale
Hierarchical Planning Units at the Regional and Planning
Area Levels
New Town Brief
Measuring Density at the Level of the Individual Site
Occupancy Rates (1986 Standards)
Land Use for a Prototype New Town of 40,000 Dwelling
Units
Site Area Standards for Residential Use and Roads by
Density
Classification of Roads and Spacing Standards
47-48
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
4-1
4-2
4-3
4-4
4-5
4-6
4-7
5-1
5-2
5-3
5-4
50-51
54
57
57
58
59
60
65
66
66
Land Use Distribution - Punggol and Pasir Ris DGPs
Planning Policies at the DGP Level and their Relationship
to National Development Objectives
Planning Policies and Performance Indicators - Punggol
DGP
Floor Area Ratio, Net Housing Yield and Population
Density
Proposed Number of Dwelling Units and Land
Consumption by Housing Type - Punggol and Pasir Ris
DGPs
Planning Guidelines for Light Rail Transit Versus Actual
Provisions in Punggol DGP
Park and Garden Provision - Punggol and Pasir Ris DGPs
74
Classification of the Variety of Housing Types
Planning Standards for Public Housing and Land
Required for Transportation Facilities
Roads - Punggol and Pasir Ris DGPs
Proposals to Reduce the Width of the Road Reserve
103
76-77
78
81
82
91
97
111
114
121
5-5
6-1
Reducing the Width of the Road Reserve and the
Resultant Land Savings
Summary of Findings, Recommendations and Questions
Arising from the Study.
122
131
Preface
Before I delve into this study, I feel it is important for me to explain how my interest in
the topic developed.
I grew up in the 1970s and 80s when Singapore was experiencing buoyant economic
growth. Despite the relative economic prosperity we enjoyed, it was impressed upon
young Singaporeans like myself how vulnerable our city-state of Singapore was, and still
is. Lacking in natural resources such as land, potable water and mineral deposits, the
story of Singapore's economic success in the 30 years since her independence is no short
of phenomenal.
The scarcity of land is very real to me.
The 1970s and 80s was a period of rapid
development. I witnessed landscapes
changing frequently before my very
eyes. I live in the eastern part of the
island. In the 1970s, my home was at
the edge of the urban area, and was
considered remote in terms of location.
Today, it is relatively close to the
Tampines regional center and several
new towns. As the urban boundary
4 Apokg
continues to shift further and further
away from my doorstep, I sometimes
wonder if there will be any land left in
time to come.
My Singapore
The growth of the urbanarea aroundmy home
in my lifetime
Economic prosperity brought about many new conveniences. For example, the travel
time by car from my home to school was reduced from 45 minutes to 20 minutes with the
completion of the East Coast Parkway (expressway) in the early 1980s. Public
transportation also became more convenient and travel times were significantly reduced
when the MRT came to my neighborhood. I am part of the generation that has grown
used to the improvements in the quality of life and certainly expect more.
Over the years, regional economic competition for foreign investment has become
keener. Singapore's neighbors have repeatedly expressed their desire to surpass her to
become the financial center, and maritime and aviation hub of the region. As the next
millennium approaches, Singapore is faced with the challenge of sustaining economic
growth, as well as enhancing the quality of life for the increasingly affluent population.
The biggest constraint is Singapore's limited land area.
It is in the light of this context that I ask the question how can the use of land as a
resource be optimized.
1
Introduction
Land is a scarce resource in the city-state' of Singapore. With an area of 648 square
kilometers 2 (250.2 square miles), the country is merely 3.5 times the size of the District
of Columbia. The demand for land is immense. Housing, jobs, infrastructure3 , education
and recreation have to be provided for within this finite space to satisfy the basic
functional needs of the 3.87 million4 people. Having enjoyed economic prosperity over
the past two decades, Singapore's increasingly affluent population expects their quality of
life to continue to be enhanced. This translates into the demand for more living space,
wider choice of housing types, and more entertainment and recreation areas. In addition,
Singapore is also faced with the challenge of sustaining economic growth in an
increasingly competitive global environment. And as a sovereign nation, Singapore has
to set land aside for the purpose of national defense and security.
With a population density5 of 6173 persons per square kilometer, land use is already
considered very intensive by world standards (Figure 1-1). Although land reclamation
has been aggressively pursed as a strategy to increase the total land area since the 1960s,
there are economic, political and environmental limits to the amount of land that can be
reclaimed from the sea. The urbanized area has been growing at a faster rate than the
territorial land area (through reclamation) (Figure 1-2). This suggests an imbalance
between the supply of land with demand. Today, less than a fifth of the existing land
area is soft land available for development in the foreseeable future . There seems to be
the real danger of land running out at some point if it is not properly managed as a
resource. Singapore's planners thus face the unenviable task of having to perform the
delicate balancing act of juggling between the various demands for land with its very
finite supply
Population and Population Density - Singapore and World Cities
Figure 1-1
GOING ON FOUR
How we've grown
-
Total. population In millions
LAND AREA
4.0
+585 sq km
(excludin
offshore islands)
.0
1980. 1990.-_ Z000
....1970.
How. we compA
1.9
.Total
population In millions
1.0
20
LENCIOF
CWTUNE
S
New York.U
*150.5 km
Sao Paulo. Brazi
15
Bombay. India
14
Sh
Shanghai, China
12;*
11
Beijing. China
10
FPULATION
Osaka. Japan
-Cairo. Egypt'
Paris. France
Jakarta. Indonesia-
+6,615 people
per sq km
(based on
present
population of
3.87 million)
+ 6,838 people
per sq km
(based on
population of
London. Britain
Bangkok. Thailand
Hongkong. China
7
6
Boston. US
5Toronto. Canada
SINGAPORE
4
Sydney. Australia
3 iLFrankfurt. Gena
Montreal Can
Rome. Italy
-Taipei. Taiwan
KL. Malaysia
4 million)
Sources: WORLD ALMANAC 98. UN
TIEN CIIUNC PINC
jr
THE PACIFIC ISLAND
OF GUAM, WHICH IS
ABOUT THE SAME
SIZE AS SINGAPORE
(541.3 SO KM), HAS A
POPULATION OF
r
148,060 - WHICH IS
I
27 TIMES SMALLER
THAN SINGAPORE'S.
wsmm-.-m---------__
Source: The Straits Times (Feb 10 1999 & Mar 20 1999)
Singaporecompares
other cities
Populatlt -Land area
(mlif In)
(sq kin)
People
(per sq kn)
Singapore
4
585
6,838
Hongkong
6
1,095
5,479.5
London
beijing
New York
7
.11
20
1,579
16,800
137,304
4,433.2
654.8
145.7
Figure 1-2
Growth of Land Area and Urbanized Area (1970 to Year X)
Land Area and Urbanized Area
800
700
600-
E400500
0
cc
440
300
200
100
0
1970
1990
1999
2010
2030
X
Year
-e-
Land Area -B-
Urbanized Area
Source: Urban Redevelopment Authority (1991)
Since attaining political independence in 1965, planning policy in Singapore has been
driven by the goal of land use optimization. Recognizing that Singapore is a city without
a hinterland, the government has actively pursued strategies of land conservation in order
to use land efficiently. Allowing development at higher spot8 densities than ever before
has been one of the main vehicles to achieve this goal. Over the last 10 years, developers
have readily arisen to the government's call to intensify site density by increasing the
floor area ratios9 (FARs) on individual land parcels upon development or redevelopment.
The previously unheard of en-bloc sale phenomenon swept through the property market.
This saw individual property owners collectively agreeing to sell their properties to a
single developer who would comprehensively redevelop the land at the higher prescribed
FAR. Although this policy has proven very successful, it is unclear if it has resulted in
the optimization of the use of land.
At this juncture, it is critical for planners and policy makers to question how effective
these strategies and policies have been in optimizing land use. Singapore's total
population will reach 4 million in mid-1999, owing to the government's pro-talent
immigration policy. Thus the population has now reached the milestone figure' 0 for Year
X in the Concept Plan" although planners thought it would be reached only after the
Year 2010. This Concept Plan which was prepared in 1991 is currently being reviewed,
and will be replaced in the Year 2001 by the new Concept Plan.
1.1
Land Use Optimization and Density
Land use optimization has always been understood as putting land to the highest and best
use. The first question which arises is how is usage of land measured? The next question
that comes to mind is what constitutes "highest" and "best"? These are relative terms
which suggest that the performance of the policies and strategies to achieve the goal of
optimization is a key consideration. The answer to both these questions is density.
Density as a construct measures the degree to which a space or area has been filled. It is
also an indicator of the performance of planning policies. Density provides a measure of
the efficient use of urban resources and the effectiveness of investment in these resources.
Density expressed as policy has a great impact on land prices, which is an important
concern of the government, especially in light of the current economic crisis affecting
Asia.
1.2
Scope
There are three different types of density, perceived density', physical density and
measured density' 3 . Each represents different phenomena and appears in different
contexts. However they are intimately linked to one another (Figure 1-3).
Figure 1-3
Relationship between different types of density
Source: Alexander & Reed (1988)
Density is a perceived experience
.
Physical density, combined with socio-cultural and
individual cognitive factors, produces perceived density. Planners and designers strive to
create environments which will positively affect people's perception of densities,
stimulating positive evaluation on the part of users.
Physical density is made up of both objective and physical characteristics of the built
environment that are perceived by people in an actual setting. Qualitative measures of
physical density include height, spacing between buildings, scale and color. These
attributes are usually not included in the quantitative measurement of density.
Quantitative measures of physical density try to capture the degree to which a space is
filled by people, structures, housing or even built-up floor space.
This study focuses on the quantitative measures of density, which are widely used to
regulate the built environment and impact urban form. It looks at physical density in two
ways. Firstly, density is recognized as a planning tool to capture the intensity of land use.
Secondly, density is a performance indicator of the successes or failures of planning
policies. The study will concentrate on residential density, which itself is a complex
notion. Worker (employment) densities and commercial densities will not be discussed.
1.3
Aim
The aim of this research is twofold. Firstly the study seeks to find an appropriate set of
quantitative density measures for Singapore given its unique development goals. The
conventional approach of using FAR alone to measure density will be questioned in order
to detect if the existing measures of density are appropriate and effective. It is also hoped
that this work will help planners and policy makers in evaluating the current planning
practice and determining its success in optimizing land use.
The study will contribute to the body of research on measured physical density. These
studies have been primarily confined to developed countries of the west. The
applications of this study may be useful in countries with similar development constraints
and goals as Singapore.
1.4
Methodology
The study methodology can be divided into two broad steps, background and analysis.
Background
The background research involved the following steps:
-
Cataloguing of land use planning policies in Singapore and relating them to the
development goals of land use optimization and improvement in the quality of life;
-
Establishing an exhaustive list of generic quantitative density measures; and
-
Researching the quantitative measures of density used and planning standards used in
Singapore.
The research was primarily library-based, although information on the Singapore
situation was also obtained through the interview process. Several planners in the Urban
Redevelopment Authority
(URA), Singapore's national planning agency, were
consulted for background information for this study.
Analysis
In this portion of the study, the performance of planning policies was analyzed in order to
assess their success in optimizing land use. A case study, the Development Guide Plan
(DGP) for the Punggol Planning Area, was be used to do this. From the analysis,
conclusions will be drawn on the effectiveness of the existing set of density measures,
and how planning policies have performed in relation to the goal to optimize land use.
1.5
Overview of the Thesis
Chapter 1
Introduction
The problem of development pressure on land in Singapore and the importance of
managing the finite land resource are introduced in this first chapter. It briefly deals with
the subject of this thesis, defines the scope of the study and highlights the methodology
used.
Chapter 2
Land Use Policies and Physical Planning in Singapore
This chapter provides the background by examining the key issues presented in the
introductory chapter. Firstly, Singapore's planning and land use policies will be
examined in relation to development goals and strategies. Performance indicators to
assess the success and failure of policies are introduced and elaborated upon. Secondly
Singapore's approach to physical planning will be discussed.
Chapter 3
Measuring Physical Density
Chapter 3 focuses on quantitative measurements of density. An exhaustive list of the
generic quantitative measures of density will be provided. The different ways density is
being measured in Singapore and how planning (provision) standards are used will also
be discussed.
Chapter 4
Analyzing Land Use Planning Policies
This chapter shifts to the analysis of the performance of land use planning policies. The
case study, the Punggol Development Guide Plan (DGP), is introduced in this chapter.
The planning policies will be evaluated against the set of performance indicators
discussed in Chapter 2 to assess their successes and failures.
Chapter 5
Strategies to Optimize the Use of Land
Chapter 5 explores the important issues related to the optimization of land use that have
not been addressed by the policies given in the case study. It looks in detail at two issues
- the classification of housing variety and the pattern of land consumption. In relation to
the latter, additional strategies to optimize land use will be proposed and evaluated to
assess their feasibility.
Chapter 6
Findings and Recommendations
In this final chapter, the suitability of density measurement methods used in Singapore
and how successful the current planning practice has been in optimizing land use will be
discussed. Recommendations will be made on alternative strategies to put land to more
optimal use. The potential areas for further study are highlighted and the shortcomings of
this exercise is discussed.
1.6
Limitations
Before delving into the study proper, it is appropriate to establish the limitations of this
study. They are with regards to the data used in this study, and the opinions and views
expressed here.
Data
The data used for analysis is not the most recent, and may have been superseded by
events. Nevertheless they are sufficient for the purposes of this study because they give
an idea of the processes behind planning in Singapore.
Views and Opinions
Having worked as an urban planner in Singapore for several years, I have gained a
valuable insight and an appreciation of the inside view of the planning practice. Being at
MIT has permitted me to view this practice from an outside perspective. In researching
the questions raised in this academic exercise, I have hoped to synthesize these diverse
experiences. I wish to qualify that the views and opinions expressed in this piece of work
are solely my own. They do not in any way reflect the official position on the issues
raised in the study.
document, city-state is used interchangeably with city, country and nation.
Singapore Facts and Pictures, 1997. The main island itself has an area of 585 square kilometres.
3 Includes transportation, water, power and sewerage
4 Department of Statistics, Singapore (published in the Straits Times, Feb 10 1999). This is the total
population figure as of mid-1998, and includes both residents and foreign workers.
5Total population density, which is different from resident population density
6 Soft land can be defined as land available for development in the next 30 years.
7 Urban Redevelopment Authority, 1996
8 i.e. on individual sites
9 The equivalent term used in Singapore for Floor Area Ratio (FAR) is Plot Ratio (PR). It is defined in
Section 1.1.11 of the Master Plan Written Statement 1998 as the ratio between the floor area of the building
and site area.
10 This planned milestone population figure of 4 million considers the resident population (Singapore
citizens and permanent residents) only. With the pro-talent immigration policy, the resident population has
increased at a faster rate than projected in the 1991 Concept Plan through migration.
" The 1991 Concept Plan, Living the Next Lap, is Singapore's blueprint for long-range physical
development up to the target planning year "X", where resident population will reach 4.4 million. It
articulates the vision for land use and transportation in 3 stages - Year 2000, Year 2010, and Year X. At
the time of preparation of this plan, foreign workers were mainly employed in the construction industry or
as domestic maids. Their housing needs did not have to be catered to because they live on construction
sites or with their employers respectively.
1 Rapoport, 1975
13 Alexander & Reed,
1988
14 Rapoport,
1975
1 The Urban Redevelopment Authority (URA) is a statutory board under the purview of the Ministry of
National Development (MND). This Ministry is responsible for Singapore's physical development through
long-term land use planning, public housing, public works, urban redevelopment, parks and recreation and
other aspects of the physical development of Singapore. URA has a crucial role in planning the
Singapore's future. It prepares long-range, as well as more detailed, local area plans for physical
development and then co-ordinates and guides efforts to bring these plans to reality.
1In this
2
2
Land Use Planning Policies and
Physical Planning in Singapore
This chapter sets the stage for the rest of the study by developing the key ideas presented
in the introductory chapter. Singapore's planning policies will be examined in relation to
development goals, and the performance indicators to assess the success and failure of
policies will be introduced. Figure 2-1 gives an overview of the relationship between the
development goals, planning policies and performance indicators. In later part of this
chapter, an overview of Singapore's approach to physical planning will be given.
Figure 2-1
Relationship between Goals, Policies and Performance Indicators
National Development
Goals
-
Managing land as a
resource
-
-
Land Use Policies
.
Building a land
bank for
contingencies
Optimizing the
use of land
Improving the
quality of life
- Improving the
living
environment and
providing more
leisure
opportunities
- Providing wider
choice of housing
types
Performance
Indicators
Quantitative
-
Density
Provision standards
Developable land
area
Development cost
Qualitative
-
Oualitv of life
2.1
Development Goals and Planning Policies in Singapore
The first chapter briefly introduced the primary physical planning challenges faced in
Singapore, namely land scarcity and people's expectations for an improved quality of
life. These central issues have shaped the nation's physical development goals, and
planners have formulated strategies and policies in order to achieve these goals.
2.1.1
Managing land as a resource
Land use planning policies related to managing land as a resource are guided by two
goals:
-
Building up and safeguarding a land bank for contingencies as well as to maintain
flexibility in responding to future changes in land demand pattern; and
-
Optimizing the use of land.
2.1.2
Quality of life
Policies related to quality of life issues are also guided by two goals:
-
Enhancing the quality of life through improvements to the living environment and
provision of adequate leisure opportunity; and
-
Achieving a private: public housing mix of 25 : 75 by 2010 in order to realize the
vision of providing greater choice and variety of housing and to meet the housing
aspirations of a better educated population.
Table 2-1 gives a catalog of the planning policies related to the development goals
articulated above.
2.2
Performance Indicators
Performance indicators help us evaluate and measure the successes and failures of
planning policies. The list of proposed indicators that can be used to assess the
performance of the policies is given in Table 2-1. They are divided into two categories,
namely quantitative and qualitative indicators.
2.2.1
Quantitative Indicators
Quantitative indicators measure the various attributes related to the built environment.
Four indicators are proposed.
"
Density
Density indicates how intensively land is being used for development. It is a very
important indicator of performance and the subject of study of this thesis. Density will be
discussed in greater detail in the following chapter.
Planning standards
Planning standards fall into three categories, namely provision standards, site area
standards and spacing standards. Provision standards are set at the macro level. Some
examples include park space per person, living space per person and road length per
person are examples of provision standards. An absolute minimum threshold is
determined for each land use in the attempt to quantify variables of the natural and builtenvironment which impact the quality of life. This would ensure an acceptable quality of
life for the city's inhabitants. Provision standards have a profound impact on density as
they determine the acceptable baseline governing the use of land and built space. For
example, if the minimum size of a dwelling unit is reduced (while every other variable is
held constant), then overall density can be increased as the same land area will generate a
higher net yield of dwelling units.
Site area and spacing standards relate to the individual site. They ensure that the built
environment has an acceptable environmental quality. Like provision standards, they
also affect density.
-
Land take
This measures how much land has been freed up or saved for development owing to
policy changes, e.g. locating compatible library and community center facilities together
on one plot of land rather than separate plots.
-
Development cost (monetary)
It is important to quantify the monetary cost incurred in the pursuit of certain policies
such as land reclamation, i.e. unit cost per hectare. This will help to determine the uses
programmed onto reclaimed land. The high cost of land usually warrants a high-value
use on the land, i.e. high-density housing or commercial development.
The first three indicators, density, planning standards and land take are related to one
another. The relationships are briefly explained.
-
Planningstandards(independent variable)and density (dependent variable)
When planning standards change, density will also follow suit. If provision standards
become more stringent, then density will increase (this assumes that there is no change in
land take). Hence provision standards are positively correlated with density. With
regards to site area and spacing standards, when these standards become stricter, density
will decrease. These two variables have a negative relationship.
-
Planningstandards(independentvariable) and land take (dependent variable)
Planning standards also have an impact on land take. If provision, site area or spacing
standards are raised, more land is usually consumed, though the rate may differ
depending on the degree to which the standards are modified. The correlation between
these two variables should be relatively high, and they have a positive relationship.
-
Density (independent variable) and land take (dependent variable)
Density and land take are negatively correlated with each other. When density falls, we
expect more land to be consumed. For example, low-density housing takes up more land
than high-density housing. How well correlated these two variables are will depend on a
number of factors.
2.2.2
Qualitative Indicators
Qualitative Indicators deal with human perceptions of the built environment and tend to
be more difficult to measure. Nevertheless they are very significant because they
contribute to people's overall satisfaction with their living environment.
e
Quality of life
Although it is possible to measure the number of sports, recreation, leisure and cultural
facilities provided, it is difficult to determine how they actually contribute to enhancing
the overall quality of life. Besides individual perceptions as to the definition of the
quality of life may differ vastly from one another. This depends on individual experience
as well as socio-cultural factors.
Environmental quality
Although this is a very important variable, it is difficult to measure its impacts for several
reasons. The impact of development may not be fully felt till several years after
completion. The signs to look out for may not even be known. Finally nature may be
able to adapt and respond to changes in the environment, e.g. migratory birds nest at the
vacant reclaimed land at Marine Parade and Marina East as their traditional habitats
elsewhere on the island have been destroyed.
-
Time Sensitivity
It is important to assess if policies can adapt to changing demands for land over time
especially in the light of the aging population. However there is no clear- cut way to
measure if a policy incorporates an in-built mechanism to respond to changes brought
about by the passage of time.
Table 2-1
Development Goals and Planning Policies
National Development
Objectives
LAND SCARCITY
Planning strategy
Planning Policies
Conserve land in the following ways:
-safeguarding large land banks
Land take;
Environmental quality
-developing substantial pockets of highdensity areas
Density;
Environmental quality
-intensifying developments in other areas
Density;
Environmental quality
Land take;
Density
Planning standards;
Land take;
Development cost;
Density;
Environmental quality
Development cost;
Environmental quality
Land take
Select sites for implementation of creative and
economic uses
Draw up planning guidelines for underground
development
Reclaim land from the sea for development
Study ways to improve the flexibility for
private sites to respond to changes in the
demand for land
Optimization of land
use
Performance Indicators
Maximize development potential
Sites around MRT/LRT stations
Spaces above roads
Density
Land take;
Density
Minimize development constraints, e.g. height Land take;
constraints, nuisance buffers around utility
Density;
Planning standards;
sites
________________________________Environmental quality
Land take;
Integrate community facilities such as
Density;
community centers, cultural institutions,
Planning standards
schools, parks etc.
Land take;
Identify new ideas to optimize land use with
Density
conservation buildings
Land take;
Identify soft land for suitable interim uses
Environmental quality;
Time- sensitivity
Land take;
Come up with incentives for new
Density;
developments and redevelopment which the
Time-sensitivity
government wants to encourage
Land take
Make use of incidental space, i.e. space under
roads
QUALITY OF LIFE ISSUES
Housing Mix by Year
Review target housing mix
2010 (25:75)
Quality of life
Create an entertainment district to add
improvement
vibrancy to night life
Introduce more street life to historic districts
Develop Seaside recreational areas
Encourage river activities
2.3
2.3.1
Quality of life;
Land take;
L '6ensity
Quality of life
Quality of life
Quality of life;
Environmental quality
Quality of life;
Environmental quality
Physical Planning in Singapore
Background
Sir Stamford Raffles of the British East India Company founded Singapore in 1819.
During the colonial era, land use planning was largely ad hoc, with the exception of the
1822 Jackson Plan'. Since independence in 1959, it has evolved to become a systematic
programmed exercise2. The government initially adopted a blueprint approach 3 in
planning resulting in the development first of the Master Plan in 1958. This was
followed by the process planning approach 4 which gave rise to the Concept Plan in 1971.
2.3.2
Evolution of the Planning System
The Master Plan
The Master Plan is a statutory plan which governs how land should to be used and the
intensity of its development. It comprises a set of land use/density maps and an
accompanying Written Statement . In accordance with the Planning Act (Number 3)
1998', the Master Plan has to be reviewed once every 5 years by the competent
authority 7 . The plan functions primarily as an instrument of control to manage the
growth and development of the city. Inevitably, it also has a significant impact on the
land values.
The first Master Plan was adopted in 1958, the last year of colonial rule. The Master
Plan has since undergone seven revisions, in 1965, 1970, 1975, 1980, 1985, 1990 and
1998. Singapore's eight Master Plans have been prepared under the auspices of three
different national governments. The first was prepared under Colonial rule, the second
under the Federation of Malaysia , and the other six under the Republic of Singapore , an
independent state. The plans under the different governments reflected their different
ideologies and political agendas.
Post-independence land use planning is characterized by its orientation towards goal
achievement.
It was not confined to influencing spatial development alone. Planning
policies were underpinned by a specific set of principles: a vision of Singapore as a
global city, and the notions of national survival and achievement. Land use planning was
integrated into a wider national policy which saw massive public investment in the
pursuit of goals that were egalitarian and aimed at ensuring the optimal use of land. Land
was treated as a scarce resource, land use was prioritized and the overall planning goals
were expressed in national rather than communal or individual terms. In the planning
process, nothing was left to chance. The preparation of the Master Plan involved
exhaustive studies and surveys being carried out to test various feasibility scenarios and
cater to contingencies.
During the period following independence, the government realized the shortcomings of
the Master Plan. Being a blueprint for land development, the primary problem was its
rigid prescriptive nature. There was the danger of strictly adhering to the blueprint rather
than adapting to the day to day changes brought about by the dynamic conditions
affecting the economy. In addition, the silent assumptions of slow and steady rate of
urban growth and social change were overthrown by the high rate of population growth.
Rapid changes made the Master Plan obsolete as soon as it was completed. The
government also realized that it could not rely on private sector initiative in physical
development to provide massive infrastructure requirements in the early days of
independence. To overcome the weaknesses of the Master Plan, Singapore undertook the
State and City Planning project" in 1967 to prepare a long-range comprehensive physical
development plan. The result was the Concept Plan.
The Concept Plan
The Concept Plan serves as a comprehensive physical development policy expressing
how land should be used in the long-term. It sets out guidelines for the optimum spatial
distribution of major activities, a framework for integrating land use, infrastructure
provision and transportation planning, and the provision of a rational basis for which the
Master Plan is to be revised periodically in order to regulate urban development. Unlike
the Master Plan, it is advisory in capacity.
The first concept plan, the Ring Concept Plan was a far-sighted plan for industrialization
and economic growth (Figure 2-2). It focused on the key infrastructure to be developed,
primarily the new airport at Changi, new residential towns, a network of expressways,
and the Mass Rapid Transit2 (MRT) system. The target planning population was 4
million, which was expected to be reached by the early part of the
2 1st century.
The second concept plan (1991), Living the Next Lap, is more comprehensive in nature
(Figure 2-2). It spells out how economic growth can be sustained, how best to plan the
transportation system, how to maintain the Asian character of the city and how to
improve the quality of life. The target population for this plan was also 4 million, which
planners thought would be reached after the Year 2010.
The Urban Redevelopment Authority (URA), Singapore's national planning agency, is
currently preparing the third Concept Plan, which they expect to complete in 2001.
Figure 2-2
Singapore's Concept Plans
QoCz5
Concept Plan 1971
URBAN DISTRICTS
II JI lGII OENSITY RFSInENTIAL AREAS
Rfl tlfl INOUSTRY
MASS TRANSIT
EXPRES$WAYS
0
UIlIAN CENTnrS
.
Source: Waller (1998)
aICONAL
SOAMIG
CINTM
CENTRE
2.3.3
The Plans
The Concept Plan is the macro-level plan which sets the strategic directions for all
subsequent plans. All the growth needs of the future are identified, and typology studies
and standard setting have to be carried out to prepare the plan.
The Layer Plans are macro-level plans which look at various important elements of the
Concept Plan. There are six Layer Plans, namely the strategic transportation plan, natural
environment plan, leisure plan, culture plan, sports plan and science habitat plan.
The Development Guide Plans (DGP) are detailed plans which begin where the broad
Concept Plan ends. Singapore island is divided into 55 planning areas, each with its own
DGP. The development quantum in the Concept Plan for each land use type is
programmed into the DGPs, and the details on land use and floor area ratio for individual
land parcels are worked out. Complementing DGPs are urban design and conservation
plans. All these plans incorporate the visions and new ideas for Singapore and form the
basis for facilitating the development of land. The 55 DGPs together form the statutory
1998 Master Plan.
2.4
Planning Approaches
Planning approaches guide the preparation of plans to achieve the vision and goals
articulated for a city. Singapore has always adopted a quantitative approach to land use
planning (Figure 2-3). To match land supply against demand still remains the starting
point of any planning exercise. This approach ensures that there is land at the right places
to meet long-term growth needs. Apart from demand and supply, the variable third
element in the approach is density 3 . The three elements in the quantitative approach will
be elaborated upon in greater detail.
Element 1: Determining the demand for land
In Singapore's case, the city is considered a receptacle, in which all the needs of the
people will be contained. The size of the receptacle is small and fixed. To cater for
every need, planners have to carefully pack in all the land uses at relatively high densities
within the tiny space. The floor space quantum for every major need, with ample
contingencies, have to be projected so that adequate land is safeguarded for every known
and anticipated need.
Element 2: Determining the supply of land
The size of the container needs to be defined to test the ultimate holding capacity of the
land and sea. Land for urbanization is the net of the natural areas to be conserved from
the total land area (includes land to be reclaimed from the sea within practical limits).
Element 3: Determining the intensity of development or density
Land use typologies and their hierarchies have to be established. Typological studies, in
terms of developmental forms, intensities and heights, are carried out to examine the
range of development prototypes for Singapore. Planners have to see how much of each
type of development we can afford to have, from the lowest to highest development
intensities within the urbanized area. This allows the maximum the range of
environmental variations without fear of facing land shortage.
Figure 2-3
Singapore's Planning Approach
SUPPLY OF LAND
DEMAND FOR LAND
-
Housing
Jobs
Infrastructure
Recreation and
Leisure
Institutions
Defense
-
Developed Land
Soft Land (available
for development in the
next 20 years)
Undeveloped land
DENSITY
Develop a typology for each
land use
2.5
2.5.1
The Planning Process and the Plans
The Process
In order to understand how density is measured, it is necessary to understand how the
planning process works. Figure 2-4 illustrates briefly Singapore's planning process.
Figure 2-4
2.5.2
The Planning Process
The Plans
The Concept Plan is the macro-level plan which sets the strategic directions for all
subsequent plans. All the growth needs of the future are identified, and typology studies
and standard setting have to be carried out to prepare the plan.
The Layer Plans are macro-level plans which look at various important elements of the
Concept Plan. There are six Layer Plans, namely the strategic transportation plan, natural
environment plan, leisure plan, culture plan, sports plan and science habitat plan.
The Development Guide Plans (DGP) are detailed plans which begin where the broad
Concept Plan ends. Singapore island is divided into 55 planning areas, each with its own
DGP. The development quantum in the Concept Plan for each land use type is
programmed into the DGPs, and the details on land use and floor area ratio for individual
land parcels are worked out. Complementing DGPs are urban design and conservation
plans. All these plans incorporate the visions and new ideas for Singapore and form the
basis for facilitating the development of land. The 55 DGPs together form the statutory
1998 Master Plan.
Next Chapter
With an idea of the background of physical planning in Singapore, the next chapter
proceeds to deal with density, the subject of this study.
This plan envisioned a strict regularity in the layout of the streets and incorporated provisions for the
separation of indigenous and European inhabitants along racial and social lines. Its aim was to achieve an
ordered pattern of growth, given that Singapore was designated as a commercial and administrative center
soon after its founding. The legacy of this plan is still evident today in the street pattern in the Central
Business District (CBD), and the existence of the historic districts of Chinatown, Kampong Glam and Little
India which were based on ethnic lines.
2Teo, 1992
3Faludi, 1973. The blueprint approach works towards a predetermined, desired end state.
4 Ibid. The process approach is inherently a response-oriented, monitoring and reactive ad hoc approach to
planning decisions.
5 The Written Statement amplifies planning definitions and proposals and describes the general intent of the
Master Plan.
6 Section 8(1) of the Planning Act (Number 3) 1998 edition
7 The competent authority for the operation of the Planning Act is the Chief Planner, Urban Redevelopment
Authority, who has been appointed by the Minister of National Development under Section 3(2) of this
Act.
8 Singapore was part of the Federation of Malaysia between 1963 and 1965.
9 Singapore separated from Malaysia on 9 August 1965.
1 Teo, 1992
" This project was under the auspices of the United Nations Development Program (UNDP).
12 The MRT is Singapore's urban rail transit system, which forms the backbone of public transportation.
13 Liu,
1991
3
Measuring Physical Density
The focus of this chapter is density. Generic quantitative measures of density are
introduced to give a picture of how density is being measured in different places. The
focus then shifts to Singapore, and the different ways density is being measured and how
planning standards are used will be discussed.
3.1
Measured Density
Density as employed in land use planning and related applications appears to be a simple
concept. However the complex reality to which it is applied - the three dimensional city
- cannot be fully captured by any given density measure. Although a useful construct,
density is inadequate to describe what the existing urban environment is like, what future
development will be like, or prescribing what it should look like. Hence there are many
measures of physical density, each trying to capture different aspects of reality as
precisely as possible. However, in view of the numerous technical complexities involved
in coming up with these measures, they are frequently misunderstood and misused.
Which measure is the most appropriate to use in a given context needs to be further
examined.
Density measures the degree to which a space or area is filled by people, residential
structures or built-up floor area. Many measures exist to express density.
3.1.1
Ratios
Density is most commonly measured in the form of ratios. Occupiers of space, such as
persons, rooms, households and dwelling units, or built-up area are found in the
numerator. The denominator generally comprises a unit of area, such as plot area,
residential land area, neighborhood area or city area. Examples of density expressed in
ratios include persons per hectare, number of dwelling units per hectare and Floor Area
Ratio (FAR). Density measured in ratios will be elaborated upon in this chapter.
3.1.2
Surrogate density measures
Density can also be reflected in terms of the physical form and volume of buildings. The
envelope of a building on a site is often determined by building coverage, height and
setbacks, which are classes of measures that are surrogates for density.
Measuring Density with Ratios
3.2
This section looks at the components of density expressed in the form of a ratio.
3.2.1
The Numerator
Numerator units can be group into three categories:
Population Units
-
-
Residential Population
=
Persons
=
Families or Households
Working Population
=
Workers
Housing Units
-
Number of housing units or dwelling units
-
Number of habitable rooms
-
Number of bedrooms
-
Number of bed spaces
Built FloorArea (Square meters / Squarefeet)
m
Gross Floor Area (GFA) - the area within the envelope of the building, including the
thickness of its external walls;
= Net Floor Area (NFA) - the area within the envelope of a building after circulation
areas (e.g. atrium, lobby and staircases) and thickness of walls is excluded.
Population and housing units used in the numerator are related to one another, and their
relationships can usually be inferred. We can convert density measures from one
population unit to another, or from population units to household units and vice versa.
For example, to convert density measures from number of persons to number of
households, we need to know (or assume) the average household size. Likewise, to
convert density measured in population units to household units, we need to make
assumptions about the occupancy rates, i.e. one household occupies one dwelling unit or
5 persons occupy one dwelling unit.
3.2.2
The Denominator
The denominator unit in density ratios is a unit of area. According to Alexander and
Reed (1988), units of area can be classified into 2 categories:
Molecular
Space within the dwelling unit (not a commonly used denominator), i.e. room, area of
dwelling unit
Molar
Land base, i.e. site (net and gross), neighborhood area, suburban area, city area. This is
the commonly used denominator in determining density, and will be discussed further.
The unit of area varies in two respects. Firstly, it varies in the unit of measurement, i.e.
acres, hectares, square miles or square kilometers. Simple arithmetic conversions allow
us to convert one unit of measurement to another. More importantly, the land base as a
unit of area varies in the definition of its boundary (Figure 3-1). Land area boundaries
are defined differently in practice, resulting in a major source of ambiguity in the
vocabulary and makes comparisons difficult. Table 3-1 shows the definition of land area
boundaries, from the level of the residential site to the city scale.
Figure 3-1
The Land Base (Molar measure)
C) r4ET RE5ite4-TIAL SUE AREA
XH~L
SCHOOL
(3) NE
80M0o'oot:
(2) 6ROSS PESIPENTIAL Sm. AREA
ARE-A
4 CrrY AREA
RstUtAL
AWfA
PI A 6FAAMCC-
Source: Hitchcock (1994)
uat
Table 3-1
Defining the Boundary of Land
Boundary of the Land
Base'
Source and Definition
Net Residential Site Area
- APHA (1960)
" Hoffman (1967)
- Martin & March (1969)
" Keeble (1971)
- Alexander & Reed (1988)
" Loder & Bayly (1994)
- Hitchcock (1994)
" Cardew (1996)
The total land area devoted to residential dwellings, including private driveways,
yards, parking areas, and gardens.
- FHA (1965)
Ambiguous about whether street and parking areas are included
Gross Residential Site
Area
"
APHA (1960)
Stevens (1960)
" Martin & March (1969)
" FHA (1971)
" Cameron (1980)
" Alexander & Reed (1988)
Net Residential Site Area plus half the area of the perimeter of roads plus onequarter the area of the intersections
"
- Hitchcock (1994)
Net Residential Site Area plus one-half rights of way which abut residential parcels
- James (1967)
Net Residential Site Area plus 20' of peripheral roads
- Loder & Bayly (1994)
" Cardew(1996)
Net Residential Site Area plus unspecified pro-rated area of local and collector
roads
Neighborhood Area
* APHA (1960)
- Gibberd (1967)
" James (1967)
" Corvallis (1977)
" Cameron (1980)
" Alexander & Reed (1988)
- Mcloughlin (1991)
- Hitchcock (1994)
" Loder & Bayly (1994)
" Cardew (1996)
The total land area devoted to neighborhood uses, including residential land, streets,
neighborhood community uses such as schools, recreation and religion culture, and
neighborhood retail shopping.
Suburban Area
" Cardew (1996)
The total land area devoted to neighborhood uses, inclusive of land used for local
and collector roads, all non-residential uses (e.g. local open space, drainage
reserves, community facilities, schools), and regional uses (regional open space,
environmental protection areas, larger scale commercial, community and education
uses)
City Area
APHA (1960)
Jensen (1966)
James (1967)
Keeble (1971)
e
Cameron (1980)
- Alexander & Reed (1988)
" Mcloughlin (1991)
" Hitchcock (1994)
" Loder & Bayly (1994)
- Cardew (1996)
The entire area of the city, regardless of land use. This tends to be ambiguous as
this could mean the city's political boundary (vacant and undeveloped land may be
included), or only the urbanized or contiguous built-up areas.
3.3
3.3.1
"
"
"
Generic Density Measures in Ratios
Population and Housing Units
Density measures primarily relate to residential areas, or to the residential population of a
city or region. From a review of the literature on density measures, it has been found that
authors are in agreement that there are several generic definitions though there is no
standardization in the terminology used by the different authors. For the purpose of this
study, Alexander & Reed's (1988) definitions of density measures will be used as they
tend to be general and encompass most of the definitions made by other autho'rs.
Net Dwelling Density
Numerator
Population (persons, households) or housing (dwellings) units
Denominator
Net residential site area (see Table 3-1)
Gross Residential Density
Numerator
Population or housing units
Denominator
Gross residential area (see Table 3-1)
NeighborhoodDensity
Numerator
Population or housing units
Denominator
Neighborhood land (see Table 3-1)
City Density
Numerator
Population or housing units
Denominator
Entire of area of the city
Table 3-2 illustrates the density measures that have been developed over the years. They
are categorized according to the density measures used by Alexander & Reed (1988).
3.3.2
Built-Up Floor Area
FloorArea Ratio2 (Figure 3-2)
Floor Area Ratio (FAR) measures density at the net site level. It expresses the built-up
floor area 3 of a structure in relation to the net site area which the buildings sits on, both in
the same unit of measurement (i.e. square meters, square feet). FAR is usually expressed
as a decimal fraction, i.e. 0.5 which means that the built-up area is half that of the site
area.
FAR can be used to define the nature of development of an area comprising many land
parcels, or to control the intensity of development permitted on any given parcel of land
especially when the impact of large buildings on the surrounding environment is a
planning concern. It is a particularly useful measure of density where non-residential
uses are concerned, or where there is a combination of residential and non-residential
uses on the same parcel of land. It can also be used to measure density on residential
sites.
Table 3-2
Comparative Density Measures reflecting Population and Housing
Units
Net Residential
Site Area
Denominator Units
Gross Residential Neighborhood
Site Area
Area
City Area
Net Dwelling
Density
Density
Gross Residential
Density
Measures
Neighborhood
Density
City Density
APHA (1960)
Gibberd (1967)
Gross
Neighborhood
Density
Numerator Units
POPULATION
Persons
Stevens (1960)
Housing Area
Density
Gibberd (1967)
Net Neighborhood
Density
James (1967)
Net Density
Jensen (1967)
Net Density
James (1967)
Gross Density
Jensen (1966)
Gross Population
Density
APHA (1960)
DistrictDensity
Jensen (1966)
Town Density
James (1967)
Town Density
Cameron (1980)
Keeble (1969)
Gross Population
Density
HOUSING
Corvallis
(Oregon League of
Cities 1977)
Rooms
Dwelling Units
APHA (1960)
DistrictDensity
APHA (1960)
Households
APHA (1960)
FHA (1971)
Net Density
Keeble (1966)
Net Residential
Density
APHA (1960)
Gross Dwelling
Density
Gibberd (1967)
Net Neighborhood
Density
Hoffman (1967)
FHA (197 1)
Gross Density
Martin & March
(1972)
Martin & March
(1972)
Hitchcock (1994)
Parcel Density
Cameron (1980)
Net Housing
Density
APHA (1960)
Net Density
APHA (1960)
District Density
Gibberd (1967)
Gross
Neighborhood
Density
Keeble (1969)
Hitchcock (1994)
Gross Residential
Area Density
Loder & Bayly
(1994)
Neighborhood
Dwelling Density
Hitchcock (1994)
Gross Municipal
Area Density
Loder & Bayly
(1994)
Urban Center
Dwelling Density
Cardew (1996)
Metropolitan
Density
Loder & Bayly
(1994)
Site Density
Cardew (1996)
Gross Residential
Density
Hitchcock (1994)
Street Density
Cardew (1996)
Site Density
Loder & Bayly
(1994)
Net Dwelling
Density
Cardew (1996)
Net Residential
Density
Source: Alexander & Reed (1988), modified
Figure 3-2
Floor Area Ratio (FAR)
I
I
I
I
I
I
I
I
I
I
I
I
I
Vero.
FAR~z*IAR
-0
* -0
0--a
FAA= I-()
0
-4
0.,#
"
0
of *s%%.* ;,
;:=1 .0
3.4
Surrogate Measures of Density
These measures will be explained in brief because they are not part of the focus of this
study.
3.4.1
Building Coverage (Figure 3-3)
The building coverage is the ratio of the area occupied by the building on the ground
floor, i.e. its footprint, to the area of the site. It is expressed as a percentage or a decimal
fraction.
3.4.2
Height and setbacks (Figure 3-4)
The combination of building height and setbacks (minimum distances between the
building and boundaries of the site) determine the mass or envelope of the building.
Figure 3-3
Building Coverage
Figure 3-4
Setbacks
STPJET
Source: Alexander & Reed (1988)
3.5
Density Measures in the Singapore Context
Chapter 2 introduced the quantitative approach to physical planning in Singapore, and the
various plans that are prepared to guide physical development. Before specifically
discussing how density fits into the quantitative planning framework, two general
measures of density, population and urban densities, will be discussed.
3.5.1
Population Density and Urban Density
Population and urban densities are two different indicators measuring how intensively
land is being put to use. The latter is especially useful in highlighting the pattern of land
consumption in urban areas for development. Population density is derived over the total
land area (i.e. the city area), whereas urban density is based on the urbanized land area of
the city only.
Singapore's population density has grown over time since 1970 (Table 3-3). However
the urban density has actually fallen over the same time period as more land is consumed
for development (Table 3-3).
URA projects that the same pattern will continue to the
Year X. This inverse relationship between the two density measures is clearly illustrated
in Figure 3-5. The land take-up rate for urban development is increasing faster than the
rate of population growth. The spread of urbanization is clearly reflected in the two
Concept Plans (Figure 2-2 refers). The implications of this pattern of land consumption
will be revisited in the later part of this study.
Population and Urban Densities (1970 to Year X)
Table 3-3
Year 4
Parameters
1970
1990
1999
2010
2030
X
586
187
(32%)
626
302
(48%)
648
375
(58%)
665
418
(63%)
703
485
(69%)
736
544
(74%)
3,016.4
2,705.1
4,000.08
3,230.5
3,853.8
3,453.8
4,0600.0
3,664.5
4,770.0
4,000.0
(persons per km2)
4819
3540
3436
4321
6173
4985
5795
5194
5775
5213
6481
5435
Urban Density (persons per km)
9456
11094
Total
8480
10768
Resident
9975
8056
8660
7761
8371
7556
8768
7353
Land Area (km2)
Total5
Urbanized
Area 6 (%)
Population ('000)
2,074.5
Totalf
2,013.6
Resident 9
Population Densi
Total
Resident
Sources: Department of Statistics, variousyears and Urban Redevelopment Authority, 1991
Figure 3-5
Population and Urban Densities (1970 to Year X)
Population and Urban Densities
12000
10000
E
o
8000
6000
CL
4000-
0
0D 2000-
1970
1990
2010
1999
2030
Year
-0-P opulation Density (Totd)
-6-Urbon Dens ity(T otd)
-8-P opJlation Dens ity(R es ident)
-)(-Urbcn Dens ity(R es ident)
X
3.5.2
Quantifying Density: How does it work?
Planning quantum figures according to land use are determined at the macro Concept
Plan level for the whole island, the 5 regions and each of the 55 planning areas (Figure 36). These numbers are then related to the individual site at the local level. This study
focuses on the two-way relationship between the planning quantums at the macro-level
and densities at the micro-levels (Figure 3-7). In this section, the relationship between
the numbers at the different levels will be highlighted.
Figure 3-6
Singapore's Planning Areas and Regions
C.
C
~
N
~I2~ ~
NOTTO SCALE
Source: UrbanRedevelopment Authority (1998)
The Quantitative Approach
Figure 3-7
THE
QUANTITATIVE
APPROACH
DEMANDFOR
LAND
SUPPLYOF
LAND
DENSITY
PLANS
PLN
QUANTUMS
METHOD
METODBY
LAND USE
3.5.3
INDIVIDUAL
SITES
DEVELOPMENT
GUIDE PLAN
CONCEPT PLAN
--
---
QUANTUMS
BY
-
FLOORAREA
RATIO (FAR);
DENSITY
Density at the Macro Level: The City, Regions and PlanningAreas
At the city level, the Concept Plan maps out the long-term land use and development
strategy. The land use distribution for living space, working space, community space,
open space and recreation, infrastructure, military use, water catchment areas and other
uses is determined from the Year 2000 and beyond (Table 3-4). The Concept Plan
intentions are then translated into the regional and local (i.e. planning area) levels to
guide development. A hierarchy of planning units is developed for different population
sizes in order to do this (Table 3-5). Commercial centers are planned to correspond to the
population size (Table 3-5), as are other land uses. It is important to note that planning
areas are equivalent to new towns in size and concept. A public housing new town is
created to be a largely self-sufficient community, with 40% of land dedicated for
housing. Table 3-6 provides the distribution of land uses by land area in a typical new
town with 44,000 dwelling units.
Table 3-4
Land Use Distribution at the City-Scale
X
1991
Year
4.77 million
Population
3.01 million
Land Use
Area (hectares)
Percentage (%)
Area (hectares)
Percentage (%)
Living Space (Housing)
Working Space (e.g.
Commerce, Industry)
Community Space
(e.g. Institution)
Open Space/Recreation
(e.g. Golf Courses, Parks)
Infrastructure
(e.g. Transportation)
Military
Central Catchment/
Reservoirs
Others'0
(e.g. Undeveloped land,
Cemetery, Agriculture,
Quarry)
7,200
12,700
11.4
20.3
15,000
19,000
20.4
25.8
2,900
4.7
5,500
7.4
4,000
6.4
7,500
10.2
7,400
11.8
10,600
14.4
12,300
5,300
19.7
8.5
10,300
5,300
14.0
7.2
10,800
17.2
400
0.6
TOTAL
62,600
100.0
73,600
100.0
Source: UrbanRedevelopment Authority (1991)
Table 3-5
Hierarchical Planning Units at the Regional and Planning Area Levels
Planning Unit
(in ascending order by land
area)
Population Size
(Dwelling Units)
Size of Commercial Centers
(m2)
LOCAL LEVEL
Precincts
HDB Neighborhoods
New Towns (Planning Area)
2,000
4,000 - 6,000
20,000 - 60,000
3,500
75,600 - 100,800
REGION LEVEL
Regions
Central Region
160,000 - 260,000
420,000
1.5 million
17 million
Source: Urban Redevelopment Authority (1991)
Table 3-6
New Town Brief (44,000 dwelling units)
Land Use
Residential
Private
Public
Commercial
Town Center
Neighborhood Center
Educational Institutions
Institutions
Parks and Gardens
Sports Facilities
Industry
Transport
Others (e.g. Infrastructure)
TOTAL
Land Area (hectares)
Percentage (%)
64.0
256.0
)42.3
24.0
36.0
62.0
16.7
85.6
9.0
55.0
93.7
54.0
) 8.0
756.0
100.0
Source: Urban Redevelopment Authority (1991)
)
8.2
2.4
11.2
1.2
7.3
12.3
7.1
3.5.4
Density at the Site Level
Different measures are used to quantify density for the different land uses found in
Singapore. Table 3-7 gives a comprehensive list of measures taking into account the
different numerators and denominators used since the inception of the 1958 Master Plan.
I will highlight the changes over time and discuss their significance.
Table 3-7
Measuring Density at the Level of the Individual Site
Land Use Typology"
1958-1990 Master Plan
1998 Master Plan
Residential
Persons per acre
(1958, 1965, 1970);
Persons per hectare
(1975, 1980, 1985)
Not applicable12
Floor Area Ratio (FAR);
Gross Floor Area (GFA) based on
allowable FAR
Residential with Commercial
use at the 1 st story only
Commercial/
Residential 3
Commercial14
Industry / Warehouse 5
Business Park16
Hotel17
Hospital and Health Care
Place of Worship
Educational Institution
Civic and Community
Institution
Residential Use
Persons per acre/Persons per
hectare
Commercial Use
Floor Area Ratio (FAR)
Floor Area Ratio (FAR);
Gross Floor Area (GFA) based
on allowable FAR
Sports and Recreation
Residential Use
The traditional aim of density control is to prevent overcrowding". Between 1958 and
1990, density was measured in terms of population units (persons per hectare 9, i.e.
ppha). This is a complicated method of measurement. For the purpose of control, the
Master Plan prescribes the maximum residential density20 allowable for development on
an individual site. However the density in ppha needs to be translated into total housing
capacity on a site. This is done by applying an assumed occupancy rate corresponding to
the size of the dwelling unit. There are many variables involved in using this measure,
namely density, number of dwelling units, occupancy rates and size of dwelling units.
-
Density (ppha) = Total number of persons in the proposal
Net residential site area2
-
Total number of persons in the proposal =
Number of dwelling units x Occupancy Rate
-
Occupancy rates (Table 3-8) serve to determine the intensity of a proposed residential
development as well as regulate the size of dwelling units (especially flats and
condominiums). Gross floor area (GFA) of the dwelling unit will be used in
determining their sizes. Occupancy rates have been revised and reduced in 1986,
resulting in a development site being able to accommodate up to 20% more dwelling
units for a given residential density2.
Table 3-8
Occupancy Rates (1986 standards)
Type of Residential
Development
Gross Floor Area (m2)
Occupancy Rate
(persons per unit)
Landed Housing
(Detached, Semi-detached,
Terrace)
No control.
GFA is the result of compliance
with all planning and
development control parameters
such as height, setbacks, building
coverage etc.
5 ppu
Flats, Townhouses,
Condominiums
<= 80 m2
81 to 120 m2
121 to 160 m2
> 160 m2
2
3
4
5
Source: Leung (1987)
ppu
ppu
ppu
ppu
= The size of dwelling units affects the proposed density (number of persons a site may
accommodate). A developer can propose 10 flat units each with an area of 162 m2,
which gives the site a capacity of 50 ppha (5 ppu x 10 units). Alternatively, if the
developer reduces the size of the flat units by 4 m2 to 158 m2, the site will be deemed
to have a capacity of 40 ppha (4 ppu x 10 units). Although the unit size is almost the
same, the proposed density is substantially different. Developers can play around
with unit sizes to keep the proposed density of development below the prescribed
maximum, or keep any increase in density above the prescribed maximum to a
minimum, while maximizing the total floor area in the development.
In addition to using population units to control density, developments were also not
allowed to exceed a FAR of 2.0 (gross). FAR was used to control the bulk of the
building.
It is evident that controlling the intensity of residential development using density (ppha)
is a complicated affair. An attempt to simplify matters has led to the use of FAR to
measure density in the 1998 Master Plan.
The main aim of FAR control is to assess the development intensity of both residential
and non-residential proposals vis-a-vis the Master Plan provision for a development site.
The Master Plan FAR control seeks to prevent the over-development of particular sites,
which create a negative impact on surrounding areas and damage the environment.
-
Floor Area Ratio =
Gross Floor Area (GFA) of building
Site Area
The 1998 Master Plan prescribes the maximum FAR allowable for a residential site. If
the area of a development site is known, one can determine the maximum GFA
allowable. For example, if a developer owns a 1000 m2 plot of land zoned for residential
use with a FAR of 2.8 (maximum), he/she knows that 2,800 m2 of residential GFA can be
allowed to be built on the site (i.e. 1000 m2 x 2.8 = 3,000 n2).
URA has a prescribed formula to convert density in ppha to FAR. This formula is used
to determine the development charge23 baseline
of a development site in order to enable
landowners / developers to be charged for exceeding the Master Plan permitted density
for the site (this excess density is allowable on the site). Density converted to FAR is
known as the equivalent plot ratio (EPR).
Equivalent Plot Ratio (EPR)=
density (ppha) x 0.0056 (density conversion factor)
For example,
If density for a plot of land is 185 ppha, then
EPR
=
185 ppha x 0.0056
=
1.036
How this density conversion factor was derived will be explained. When the intensity of
residential development was measured in terms of ppha, it was tied to a range of
occupancy rates. The lowest occupancy rate was 2 persons per unit and the size of the
unit 80 m2. This implied that one person is assigned to 40 m2 of living area. It was
assumed that 40% of the Gross Floor Area of a residential building is given to nonresidential uses (circulation area, commercial use). One person was thus entitled to 40%
x 40 m2 of non-residential area, or 16 m2. The total space entitlement per person is 56
m2 (40 + 16 m2) or 0.0056 hectares. Hence this conversion factor of 0.0056.
Non-Residential Land Uses
With regard to non-residential land uses, FAR has been consistently used as the tool to
measure density. For commercial use, the prescribed plot ratio has always been 3.0
unless stipulated in the Master Plans prior to 1998. Industrial use has always been
prescribed a plot ratio of 2.5.
Planning Standards in Singapore
3.6
Singapore's quantitative approach to planning has resulted in planning standards being
set for each land use. Planning standards include provision standards, site area standards
and spacing standards. As these were briefly explained in the previous chapter, they will
be elaborated upon below.
3.6.1
Provision standards
Provision standards stipulate the minimum number of facilities to be provided given a
certain land area, number of housing units or population size, e.g. provision standards for
parks, open spaces, sporting facilities, cultural facilities, etc. They are an attempt to
ensure an acceptable quality of life for the population. These standards are usually
expressed in the form of ratios. These ratios will be broken down into its components to
show which variables go into the numerator and denominator respectively.
Numerator
-
Housing units (number of dwelling units)
-
Facilities including sports and cultural facilities and infrastructure. (Sports facilities
include sports stadiums, swimming complexes, sea sports centers, golf courses etc.
Cultural facilities include museums, performance theatres, music halls etc.
Infrastructure includes ports/airports and utilities such as sewerage treatment works,
incineration plants, power stations.)
-
Land area (usually in the case of open spaces and parks)
Denominator
-
Population (number of persons)
-
Land area
-
Town (i.e. new town)
-
Housing units (number of dwelling units)
Provision standards are determined on a national basis (Concept Plan) as well as a new
town basis. National standards usually have population in the denominator.
Examples of national standards include open space (area) per population, sports facilities
per population, cultural facilities per population and infrastructure per population. Town
standards usually have the town or number of dwelling units in the denominator. Table
3-9 shows the planning standards for a prototype new town (40,000 dwelling units). The
provision standards for each land use and facility are broken down into the numerator and
denominator.
3.6.2
Site area standards
These standards capture the minimum plot size for different land uses and facilities such
as schools, parks, town centers etc. Table 3-9 gives a comprehensive list of site area
standards for new towns. For residential use, they are further subdivided into the
following levels based on the number of dwelling units: new town, neighborhood (highdensity), estate and precinct (medium and low density) (Table 3-10).
3.6.3
Spacing standards
Spacing standards stipulate minimum widths or distances between two physical entities.
Examples of spacing standards include road standards and building to boundary spacing
standards. They attempt to ensure that safety and residential amenity of the population is
protected. Road standards will be elaborated upon as they are important in this study.
Roads essentially perform two functions: movement and access. The road system has
been classified by estimating the importance of the movement relative to access. The
four classes are the expressway system, the arterial system, the collector system and the
local road system. At the upper end of the scale are the expressways, whose primary
function is to move long-distance traffic. At the other end are the residential streets
which primarily provide access to houses. The classification of roads based on hierarchy
of function and spacing standards is shown in Table 3-11.
Table 3-9
Land Use for a Prototype New Town of 40,000 du (High-Density)
Land Use
Provision
Standard
Numerator /
Denominator
Site Area
Standard (ha)
Total
Provision (#)
160 du/ha 5
Dwelling
-
-
250
Facility/Town
Facility/
Housing Units
29.0
6.0
1
7
71
29
42
Facility/
1.8
13
25
3.0
8
26
6.0
1
6
6
250
Residential
Residential
Land Area
(ha)
Units/
Land Area
Commercial
Town Center
Neighborhood
Center
1 / New Town
1 / 5,000 to
6,000 du
63
Educational Institution
Primary School
1 / 4,300du
Housing Units
Secondary
1 / 6,100 du
School
Facility/
Housing Units
1 / 40,700 du
Junior College
Facility/
Housing Units
Vocational
1 / New town
Facility/Town
6.0
1
1 / 10,000 du
Facility/
0.1
4
1.8
0.5
1
0.5
Institute
Institutions
14
Community
Center
Housing Units
Polyclinic
1 / 30,000 du
Facility/
Housing Units
Library
1 / New Town
0.5
1
0.5
Place of
-
-
0.2-0.4
10
4.2
-
-
0.5
14
7
1 / New Town
1 / New Town
Facility/Town
Facility/Town
10.0
3.0
1
1
10
3
1 / 6,000 du
Facility/
Housing Units
Facility/
2.0
7
14
0.3
10
3
Worship
Reserve Sites
Parks
& Gardens
District Park
Town Center
30
Garden
Neighborhood
Park
Precinct
1 / 800 du
Garden
Housing Units
Sports & Recreation
Indoor Stadium
Sports
9
1 / New Town
1 / New Town
Facility/Town
Facility/Town
1.2
3.0
1
1
1.2
3
1 / 30,000 du
Facility/
1.5
1
1.5
0.1
33
3.3
Complex
Swimming
Pool
Games Courts
Housing Units
1 / 1,200 du
Facility/
Housing Units
Industry
Roads
Utilities
TOTAL
Source: Urban Redevelopment Authority (undated)
55
88
20
60026
Table 3-10
Site Area Standards for Residential Use and Roads by Density
Prototype
Density
Typology
Number of
Dwelling
Units
Residential Site Area
Standard
Road Site Area
Standard
New Town
High-Density
40,000
250 ha
88 ha
Neighborhood
High-Density
6,000
38.0 ha
(based on 160 du/ha)
17 ha
(includes secondary
roads)
Estate
Medium & Low
Density
12,000
200 ha
85.7
(includes roads outside
development parcels)
Precinct
Medium & Low
Density
2,000
33.3 ha
13.2
(includes roads outside
development parcels)
Source: UrbanRedevelopment Authority (undated)
Table 3-11
Road
System
Classification of Roads and Spacing Standards
Road Type
Road
Reserve
No. of
Lanes
(m)
Buffer Standards
Design
Speed
Road
Capacity
Road
Spacing
(km/h)
(vehicle/
(km)
80
14,400
2.5-9.0
30
1.5-4.0
15 (> 4 stories)
hour)27
(M) 2
FUNCTION: MOVEMENT
52.9
8
45.5
6
Semi-
45.4
8
Expressway
38.6
6
Arterial
Major
45.4
8
System
Arterial
38.6
6
Arterial
38.6
31.8
6
4
50-60
4,500
3,000
0.9-1.2
10 (> 4 stories)
7.6 (up to 4 stories)
31.8
26.2
4
4
50-60
-
0.6-0.9
7.6
21.4
2
18.0
15.4
14.2
12.2
11.6
11.1
2
2
2
2
2
2
40-50
-
0.1-0.6
2
Expressway
Expressway
System
10,800
60-70
9,600
7,200
60-70
6,000
12 (up to 4-stories)
1.2-1.5
15 (> 4 stories)
12 (up to 4stories)
4,500
FUNCTION: ACCESS
Collector
System
Primary
Access Road
Local Road
System
Local
Access
Source: UrbanRedevelopment Authority (1992)
Roads should have an adequate number of traffic lanes to handle the desired volume of
traffic. The standard lane widths are as follows:
-
3.7 meters for all lanes of an expressway, curbside lanes of a dual carriageway and all
lanes of a one-way street in the city;
= 3.4 meters for inner lanes of a dual carriageway arterial;
-
3.0 meters for local access road (except industrial areas).
Figure 3-8 shows the width of lanes for expressways, major arterial roads and arterial
roads.
Figure 3-8
Width of Travel Lanes
p- EXPRESWAY 45.%-
4
f-MMA TERIAL a6-6m-I'
HARTERIAL PI'A-a
Source: Urban Redevelopment Authority (1992)
It is important to note that the provision, site area and spacing standards have all changed
over time. The changes reflect the need to balance the demand for land with its supply.
However this study was unable to trace the actual changes in these standards due to the
limitations on data availability.
Next Chapter
Having established an exhaustive list of the generic quantitative measures of density, and
discussed how density is being measured and planning standards are used in Singapore,
the following chapter proceeds with the analysis of the planning policies that were
introduced in Chapter 2.
The terms used here are from Alexander & Reed (1988).
is also known as Plot Ratio or Floor Space Index. In Singapore, the terminology used is Plot Ratio
(Gross).
3 The built-up floor area is commonly the gross floor area (GFA) of a building.
2 FAR
4 Population figures (total and resident) for the Years 1970, 1990 and 1999 are from the Department of
Statistics. Population figures (total and resident) for the Years 2010, 2030 and X are 1991 Concept Plan
projections from the Urban Redevelopment Authority (1991).
5 Total Land Area = Living space, Working space Community space, Open space and Recreation,
Infrastructure, Military use, Catchment areas and Others (Agriculture, Quarry, Cemetery and Undeveloped
land).
6 Urbanized Land Area = Total Land Area - Catchment areas, Military use, Agriculture, Quarry, Natural
open space, Green separators, Cemetery and Undeveloped land.
7 Total Population = Resident + Non-Resident Population
8 This is the actual population figure, which exceeds the projected total population (3,585,000) in the 1991
Concept Plan. This points towards the trend where the total population will exceed the projected figures in
the Years 2010 and X.
9 Resident Population only, i.e. citizens and permanent residents. This does not include foreign workers
such as construction workers, domestic maids.
10The land area designated for land uses in the "Others" category shrinks significantly between 1991 and
Year X. As the land bank is drawn upon, the undeveloped land will be converted to other land uses, thus
contributing to the significant fall in the land area.
" From 1998 Master Plan Written Statement. This typology includes only selected land uses for the
purpose of this exercise.
12 This zone did not exist in the earlier Master
Plans.
13 The commercial quantum is restricted strictly to the lower stories. It must not exceed
40% of the total
allowable GFA based on allowable plot ratio. The residential quantum must not be less than 60% of the
total allowable GFA based on allowable plot ratio.
14 Includes Office, Mixed Commercial (office, shopping, cinema and hotel), convention
center and food
center/restaurant uses.
15 The industrial/warehouse use quantum must not be less than 60% of the total GFA. Not
more than 40%
of the total GFA should be used for other permitted ancillary, related or compatible uses.
16 At least 60% of the total GFA shall be used for any combination of business park operations.
Not more
than 40% of the total GFA should be used for other permitted ancillary uses.
17 At least 60% of the total GFA shall be used for hotel room floors. Shopping and hotel related
uses may
be considered and use quantum shall not exceed 40% of the total GFA.
18 Leung, 1987: 107
19 Residential density used to be measured in persons per acre (ppa) prior to 1974, before Singapore
switched to the metric system of measurements.
20 Maximum residential density varies between 125 ppha to 1485 ppha, depending
on location.
21 The net residential site area comprises the area of the site proper and any land to be
surrendered to the
State for road widening and open space, land for electric substation but excludes land reserved for future
development and plots to be occupied by non-residential development.
According to Leung (1987), the 20% increase in the number of units at any given density is the result of
two policy changes. Firstly, unit sizes were measured in Gross Floor Area (GFA) terms instead of Net
Floor Area (NFA), i.e. balconies and terraces were included in the floor area computation (numerator
became a bigger number). At the same time, occupancy rates prescribed for the various unit sizes were
reduced (denominator became a smaller figure). The net result was that each development site could
accommodate 20% more units for a given residential density.
23 Development charge is a betterment levy which attempts to achieve equity by recouping to the State part
2
of the community created value consequent to the grant of planning approval.
24 The baseline for development charge is the density stipulated in the 1958 or 1980 Master Plans (i.e. 185
ppha), or the approved density or FAR for which development charge has already been paid (Section 36(1)
of the Planning Act 1998 refers).
25 Each person is entitled 56 m2 of space.
The land areas for this prototype new town differs from that in Table 3-6 by 156 m2. The difference can
be attributed the inclusion of private housing area(s) as part of the town, and the tripling of the area of open
space proposed as compared with the prototype. The increase in open space area is probably due to the
green belts around the DGPs being included as part of the land base.
27 Number of cars that can traverse a point per hour. Capacity depends on many factors, i.e. alignment,
grade, surface conditions, lane width etc. A rough guide of capacity for different road types is given below:
1800 vehicle/hour/lane
Expressway
1200 vehicle/hour/lane
Semi-Expressway
600-900 vehicle/hour/lane
Major Arterials/Arterials Traffic volume not important because these roads provide access
Access (Primary and Local)28 The buffer is a strip or an area of land which falls between the limit of the road reserve and the building
line. Buildings cannot be built within this buffer. There are two components to the buffer - the physical
buffer and the green buffer. The physical buffer can be paved over to accommodate car parks, driveways,
cycle tracks and small ancillary structures like bin centers. The green buffer has to be strictly reserved for
tree planting and turfing. The standards given here are for residential land.
26
4
Analyzing Land Use Planning
Policies
This chapter introduces and discusses a case study to assess if land use planning policies
meet overall development goals. The planning policies are articulated in the form of a
plan, which will be analyzed. How these policies perform will be reflected through the
indicators of density, planning standards and land take.
4.1
Case Study - The Development Guide Plan for Punggol Planning Area
The Development Guide Plan' (DGP) for the Punggol Planning Area has been selected
for study. The Urban Redevelopment Authority (URA) and the Housing and
Development Board (HDB) have earmarked Punggol 21 as the model for future new
towns, offering a new lifestyle for the
2 1s'century.
Changes from the characteristics of
existing new towns can be expected. For instance, the Punggol DGP proposes FARs of
3.0 to 3.4 for plots of land designated for public housing use, much higher than the FAR
of 2.8 in existing housing estates. In view that the new towns of the future will be
planned and developed based on this model, it would be particularly appropriate to assess
if Punggol's land use policies meet development goals.
4.1.1
Location and Background
The Punggol Planning Area is located in the North-Eastern region of Singapore (Figure
4-1). The 957-hectare site is bounded by Sungei2 Serangoon to the east, the Tampines
Expressway (TPE) to the south, Sungei Punggol to the west, and the Straits of Johor and
Serangoon Harbor to the north. Most of the land is currently undeveloped as existing
poultry and pig farms were phased out in the 1970s for redevelopment, and the seafood
restaurants and boatels relocated to facilitate reclamation of the coastline. Construction
of the new town commenced in 1998.
4.1.2
Vision
Punggol 21 was conceived as a waterfront town, offering frontage to the water along its 3
rivers3. It will be home to 85,800 families. The key ideas for Punggol 214 are as follows:
-
More executive condominiums5 and private housing (landed housing and
condominiums ), which will comprise 10% and 30% of the proposed housing stock
respectively;
-
More high quality HDB flats;
-
A Light Rail Transit (LRT) system planned right from the start, with walking
distances not more than 300m for most residents;
m
Clustering of community facilities; and
-
A variety of clubs7 within the town.
4.1.3
Making Comparisons
For the purpose of making comparisons between existing new towns and Punggol 21,
Pasir Ris DGP8 will be used as a foil for comparison with the case study where
appropriate. The Pasir Ris planning area is located in the eastern region of Singapore and
will be home to 46,400 families9. It lies east of Punggol and they share a common
boundary along the Sungei Serangoon (Figure 4-1). Within the planning area is a new
town that was developed and occupied at the turn on the decade, around the same time
the 1991 Concept Plan was released. Pasir Ris is a suitable candidate for comparison for
two reasons:
Figure 4-1
Location of Punggol Planning Area and Pasir Ris Planning Area
SHADED AREAS ENLARGED
ILAT JOHOR
Sungei Punggol
Island
Serangoon
SERANGOON HARBOUR
TAMPINES EXPRESSWA'
*
N
'Coll
Source: Urban Redevelopment Authority (1998)
0METER
The residential land use profile is very similar to Punggol's in terms of land area, and the
type of dwelling units, which tend to be at the higher-end of the public housing1 scale.
The two towns are also similar in terms of their population profiles. Pasir Ris is home to
the new generation of Singaporeans who are more affluent than their parents, and seek a
better quality of life. Punggol 21 will similarly be home to this same segment of the
population.
-
The town was planned in a different era, and hence it would make it easy to elicit the
changes in the land use policies and provision and site area standards.
4.1.4
Land Use
Punggol is a residential town with public and private residential housing proposed. Pasir
Ris has similar land uses as Punggol with the exception of industrial use. The land use
distribution for both DGPs is shown in Table 4-1. For the purposes of this study, the
industrial land and tracts of residential land which are subject to detailed planning will be
excluded from the study area of Pasir Ris DGP. Figure 4-2 shows the area of study for
Pasir Ris DGP.
Table 4-1
Land Use Distribution - Punggol and Pasir Ris DGPs
Land Use
Land Area
(ha)
Residential
474
Punggol
Land Area
(%)
50
Land Area
(ha)
Pasir Ris
Land Area
(
530 (overall);
35 (overall);
318
-
Commercial
Industrial
Institutional
Open Space & Recreation
Road & Infrastructure
Waterbody
Others
13
0
50
131
150
114
25
1
0
5
14
15
12
3
13
376
65
127
199
67
156
1
25
4
8
13
4
10
Total
957
100
1,533 (overall)
100
882 (within
study area)
Source: Urban Redevelopment Authority (1995 & 1998)
~.
Figure 4-2
El
U
Study Area for Pasir Ris DGP
IA
STUD
'IAE
ECUE
10
4.2
AREA
W
M E
Planning Policies
The planning policies for Punggol and Pasir Ris DGPs have been mapped out in relation
to national development goals and broad planning policies (Table 4-2). In the case of
Punggol, all of the policies articulated in the DGP are spelled out at the broader national
level except for the strategy to provide an LRT system that will be integrated with the
MRT system. The same is true for Pasir Ris as all DGP policies can be tied in with
broader policies except for the strategy to provide better road networks and north-south
transportation linkages. However Punggol has more strategies that match the broad
planning policies (in terms of number). This is not surprising since the plan is much
more recent, and the strategies used are in tandem with the current trend of thinking.
Although we see some change in planning strategies over time, policies to intensify land
around transit stations, introduce a wider variety of housing and provide accessibility to
recreational areas through the green connector system are not new since they are reflected
in both DGPs.
Table 4-2
Planning Policies at the DGP level and their Relationship to National
Development Objectives and Policies
National
Development
Objectives1
SCARCITY OF LAND
Planning strategy:
To build up and
safeguard a land
bank for:
Contingencies
(e.g. increased
population);
- -flexibility to
respond to
future changes
in land demand
pattern
Planning Policies
National-scale"
Conserve land in the following
ways:
-safeguarding large land
banks
-developing substantial
pockets of high-density areas
-intensifying developments
in other areas
Select sites for implementation
of creative and economic uses
Draw up planning guidelines
for underground development
Reclaim land from the sea for
development
Punggol DGP14
Pasir Ris DGP
5
-
Increase land area
through reclamation
Study ways to improve the
flexibility for private sites to
respond to changes in the
demand for land
Optimization of land
use
Maximize development
potential
-
-
Sites around MRT/LRT
-
stations
around MRT and LRT
stations
employment centers
near the MRT station to
capitalize on its mass
transit accessibility
Introduce cluster
community facilities as
focal points for
community activities
-
Intensify land use
-
Spaces above roads
Minimize development
constraints, e.g. height
constraints, nuisance buffers
Integrate community facilities
such as community centers,
cultural institutions, schools,
parks etc.
Identify new ideas to optimize
land use with conservation
buildings
Identify soft land for suitable
interim uses
Come up with incentives for
new developments and
redevelopment which the
government wants to
-
-
-
encourage
Make use of incidental space,
i.e. space under roads
-
Provide shopping and
HousingMix by
Year 2010 (25:75)
Review target housing mix
Introduce greater
variety of housing types
Quality of life
Create an entertainment
district to add vibrancy to
night life
Introduce more street life to
historic districts
Develop Seaside recreational
areas
-
Maintain the variety of
housing types through
provision of low,
medium and high
density housing on
vacant lands when
Loyang Industrial
Estate is phased out
improvement
-
Introduce thematic
clubs and marinas;
Retain Coney Island
(with a nice natural
beach) as a Regional
(Note: Pasir Ris Park
has a stretch of sandy
beaches. Holiday
chalets are located
along the beaches).
Park;Note:PasirRisPark
Encourage river activities
4.2.1
Provide a network of
green connectors and
promenades along the
rivers
Provide greater
accessibility to the
major recreational area
via park connectors
along the rivers and
other waterbodies
Planning Policies and Performance Indicators
To set the stage to assess the performance of planning policies articulated in the case
study, it is necessary to match the performance indicators discussed in Chapter 2 with
these policies. Table 4-3 shows these performance indicators in relation to the planning
objectives and policies for the Punggol DGP. In this analysis, policies that are tied to
three of the quantitative performance indicators, density, planning standards and land
take will be further studied. 5 out of 9 of the planning policies can be assessed in terms
of these indicators, which overlap in some cases.
Table 4-3
Planning Policies and Performance Indicators - Punggol DGP
Planning Objectives 16
Planning Policies"
Performance
Indicators
Land Use
Meet Concept Plan provision of
land uses and quantum for
planning area
Increase land area through
reclamation
Intensify land use around
MRT and LRT stations
Development cost;
Environmental quality
Density;
Land take;
Environmental quality
All
Create a high-quality residential
town which will serve as a model
Introduce a greater variety
of housing types
Density;
Planning standards;
Residential
town in the
2 1s" century
Create an environment which
fosters a sense of community
bonding
Plan for a good transport
network and public transport
system to serve the
developments in the town
Provide a wide range of
recreational facilities for the
residents
Capitalize on the area's natural
attributes of riverine and coastal
features
Residential
Land take;
Introduce cluster
community facilities as
focal points for community
activities
Provide an extensive LRT
system integrated with the
MRT system
Introduce thematic clubs
Introduce marinas
______________
Retain Coney Island as a
Regional Park;
Provide a network of green
connectors and promenades
along the rivers
Quality of life;
Environmental quality
Density;
Planning standards;
Quality of life
Institutional
Density;
Planning standards;
Land take;
Quality of life
Quality of life
Transport
Quality of life;
Environmental quality
Sports and
Recreation
Quality of life;
Environmental quality;
Quality of life;
Environmental quality;
Planning standards;
Park and
Garden
Park and
Garden;
Open Space
Sports and
Recreation
Land take
4.3
Planning Policies : Evaluating Successes and Failures
Do planning policies for Punggol DGP meet the stipulated planning objectives? This
section examines those policies which have density, provision standards and land take as
indicators of performance. Although this study focuses on density, it is important to
consider provision standards and land take because they have clear relationships with the
variable, density. The successes and failures of the policies are evaluated to see if land
use has been optimized. For the purpose of analysis, the policies have been related to the
different land uses proposed in the plan (see Table 4-3).
4.3.1
Residential Use
Policy 1 - Intensify Land Use aroundMRT andLRT Stations
This policy directly addresses density, which measures how intensively land is used. For
the Punggol planning area, land use is intensified through the increase in the FAR for
sites around public transit stations. These sites are primarily zoned for residential use
with designated maximum FARs between 3.0 to 3.4. In the case of Pasir Ris DGP 8 ,
high-density residential lands are designated a FAR of 2.8. The proposed FARs for
Punggol exceed those for 2.8 Pasir Ris (Figure 4-3). Superficially, it appears that by
increasing the FAR of individual sites, land use has been intensified. To assess the real
impact of this spot FARs increase, it necessary to look at the relationship between FAR
and the net housing yield, as well as FAR and residential land consumption.
-
FAR and net housing yield
With increases in spot FARs, it follows that the number of dwelling units on residential
land increases. What is the impact of an increase in FAR on the number of housing units
that land can yield? Net housing yield (Number of dwelling units proposed / Site Area),
measured in dwelling units per hectare, illustrates the performance of each FAR range
and will help to determine the suitability of these prescribed ranges.
Table 4-4 shows the relationship between FAR, net housing yield and population density
based on URA's typology for residential use. The net housing yield shown here will be
the control figure which actual yields will be benchmarked against.
Figure 4-3
Punggol
Pasir Ris
Landed Housing
1.4
1.6
2.1
2.8
3.0
3.4
3.5
400
200
D
100
300
500 METER
Floor Area Ratios - Punggol and Pasir Ris DGPs
Table 4-4
FAR, Net Housing Yield and Population Density
Housing
Type
Residential
Typology
Density
Floor Area
Ratio
Landed20
Low Density
NonLanded21
Net Housing
Yield
(du/ha)
Population
in persons
per hectare
No FAR
designated
30-50
-
Low Density
Medium Density
1.4
1.6
2.1
80
91
120
250
285
370
High Density
2.8
160
500
3.0
3.4
3.5
171
194
200
>500
Building Height
(Stories)
(ppha)_
2-3
(depending on
landed housing
estates height
control)
4
10
16 or technical
height control,
whichever
is
lower
20 or technical
height control,
whichever is lower
Assumptions
=
Site Area
Occupancy Rate =
Gross Unit Size =
>20 of technical
height control,
whichever is lower
10,000 m2
3.11 persons per unit (ppu)
175 m2 (56 m2 x 3.11 ppu)
Source: UrbanRedevelopment Authority (undated)
Table 4-5 compares the net housing yield for each housing type in the Punggol and Pasir
Ris DGPs to the control figures in Table 4-4. In general, the proposed housing yields
match the theoretical figures quite closely. This suggest that land is being used optimally
according to URA's definition of "optimal". Nevertheless deviations form the control
figures are to be expected.
For Pasir Ris DGP, the net housing yield for each FAR category does not deviate more
than 17% except for FAR 1.4 and landed housing. The deviation for FAR 1.4 is 81% and
53% for landed housing. The deviations in both these categories (low-density) are large
because 83% of the stock of units (1665 dwelling units) has already been developed.
These units were subject to more stringent22 site area and spacing standards when
developed, resulting in more land being consumed than would have been the case if they
are developed under prevailing standards. In the case of Punggol DGP, the net housing
-1-- -ift
Eft-
Proposed Number of Dwelling Units and Land Consumption by
Housing Type - Punggol and Pasir Ris DGPs
Table 4-5
Residential
PASIR RIS
PUNGGOL
FAR
Dus
Typology
Land
area
(ha)
Net housing yield
dulha
Control Prop
Dus
Land
area
(ha)7
Net housing yield
(du/ha)
Control Prop
(P-C)
(P-C)_
Low
1000
N.A.
32
30-50
Low
1,100
1.4
20
80
Subtotal
1.6
2,100
52
(2%)
(11%)
2,500
27
-
91
80
30-50
55
40
92
8,200
2.1
_
74
120
110
14
(-16)
900
61
80
15
(-66)
2,000
141
(4%)
(31%)
3,500
36
(+1)
Density
7,900
70
-
14
91
88
'
(-3)
120
113
(-7)
_(-10)
10,700
(13%)
67
(14%)
-
105
11,400
(25%)
106
(23%)
-
114
2.8
2,400
15
160
160
(0)
33,000
205
160
160
(0)
>2.8
71,000
340
171-200
208
-
-
Subtotal
High
Density
1,100
(-25)
Density
Medium
28
(-2)
Density
-
(8-37)
Subtotal
73,000
(85%)
355
(74%)
-
206
33,000
(71%)
205
(46%)
-
160
TOTAL
85,800
474
-
181
46,400
45226
-
120
(100%)
(100%)
(100%)
(100%)
Source: Urban Redevelopment Authority, 1995 & 1998
yield for each FAR type is within a +/- 17% from the control, except for 1.4 FAR in the
low-density category. The proposed number of units is 31.25% below the control figure.
This figure supports the recent claims made by professionals in the real estate industry
who have argued that the FAR of 1.4 is unrealistic because it under-utilizes land.
However it is important to note that sites with low FARs of 1.4 and 1.6 tend to surround
landed housing areas. Sensitivity to the amenity of these landed housing estates requires
the heights and densities of the surrounding land parcels to be regulated with care. The
1.4 FAR type should be reviewed to determine if it should be revised.
Net Housing Yield - Proposed (Punggol and Pasir Ris DGPs) and
Figure 4-4
Control by Each Floor Area Ratio Range
FAR and Net Housing Yield
350-
300
0
0
250
200
0
.E150-
-
100
00
z
50-
0
i 10
rh
(
-
V
V
CII J C\i I
0
-JI
OIJ
C~i
CIIJ (%J~ CIIJ CIIIJ
C6
6~
C6
6~
C
FAR
---
Control
0
Pungl
A
PairRIs
Policy 2 - Introduce a Greater Variety of Housing Types
This policy statement does not clearly articulate what constitutes variety. However from
the vision statement for the town in the Punggol 21 report, variety in the housing stock
can be achieved in two ways - by land tenure, and by design quality.
Variety is measured by land tenure through the differentiation between private and public
housing units (Figure 4-5). Punggol will have more executive condominiums and private
housing (landed housing and condominiums), which will comprise 10% and 30% of the
proposed housing stock respectively, i.e. private housing stock is 40%. Benchmarked
against the projected public : private housing mix of 75 : 25 for the whole country in the
Year 2010, this plan has clearly surpassed the target.
Within the public housing sub-category, variety is achieved in terms of the design quality
of the housing units. Punggol will have standard HDB flats27 , design plus flats 28 and
design and build flats29 (Figure 4-6). Each type of flats will have a one-third share of the
number of public housing units. This category appeals to people's perceptions of the
quality of the built environment. As public housing still carries a stigma, and involving
private sector architects in the design on HDB flats can be one way of improving the
image of new towns and give residents more choices in the types of housing. This is one
way to meet the rising aspirations of the population. The ultimate goal should be to have
public and private housing in new towns, and housing with good design to make it
difficult to differential between the two.
Although it appears that a good variety in housing mix has been achieved, it is important
to note that variety can be defined in more ways than these two. How variety is classified
has impact on land consumption. This will be explored in the next chapter.
Figure 4-5
Housing Variety by Land Tenure
Landed Housing
Private Condominium
Executive Condominium
44e
Public Housing
Figure 4-6
Housing Variety by Design Quality
(Public Housing)
Standard Flats
Sengkang
Design Plus Flats
Choa Chu Kang
Design and Build
Flats
Choa Chu Kang
4.3.2 InstitutionalUse
Policy 3 - Introduce cluster community facilities asfocal pointsfor community activities
The Punggol DGP proposes the clustering together of community facilities such as
schools, places of worship, recreational and social facilities (clubs). Commonly, the
school field is shared between the different institutional or recreation uses. Figure 4-7
shows an example of the cluster concept, with a school field being shared by a school and
a club. Figure 4-8 illustrates the distribution of community facilities in both Punggol and
Pasir Ris DGPs. Nine clusters of community facilities are proposed for Punggol and
none in Pasir Ris. This is because the clustering of community facilities is a new policy.
Clustering appears to be an innovative
Figure 4-7
method to optimize the use of land in
two ways. Firstly it recognizes that land
A
use intensity can be expressed in more
ways that just the form of density. It
takes into account how frequently land is
used over different periods of time, i.e.
day and week. Different groups of
people make use of land at different
times of the day and week. For example,
Source: UrbanRedevelopment Authority (1998)
school buildings are occupied in the day and community centers primarily used at night.
Schools are used on weekdays and places of worship on weekends. Site area standards
can be reduced when facilities are shared between these different uses. This results in
land being saved. For example, the combined area for a secondary school and club can
be reduced by the size of one football field because a school and club share this field.
However the sharing of facilities will result in additional load on these facilities and
increase maintenance costs. Management of the facilities may be difficult. Hence the
benefits of sharing have to be weighed against the costs in order to assess the viability of
this strategy.
88
I.
_
Figure 4-8
Community Facilities - Punggol and Pasir Ris
DGPs
Punggol
/1&
4a
Pasir Ris
Primary School
-w
Secondary School
Special School
U
Civic/Community
Club
Cluster
M
wmu
4.3.2
Transportation
Policy 4 - Provide an extensive LRT system integratedwith the MRT system
Punggol new town will be served by one MRT station along the North-East Line, which
will connect the town to the rest of the island. Figure 4-9 shows the public transportation
infrastructure. The MRT station is proposed to have a catchment radius of 400 meters.
An LRT system is proposed for internal circulation within the town. The line is 19
kilometers in length, with two loops that serve the main housing areas and one spur line
which serves the residential and recreational nodes north of the new semi-expressway.
19 LRT stations are proposed, and will be located within a radius of 300-350 meters from
most homes.
Figure 4-9
Public Transportation Network - Punggol DGP
Loop 1
"
MRT Station
-
LRT Station
0
Source: Urban Redevelopment Authority, 1998
.
Na1mER
Density is important for two reasons. The high FARs proposed for residential sites
around LRT stations justifies the provision of the LRT system, which is very much more
expensive30 than the traditional feeder bus network that provides the internal
transportation service between the town center (MRT station) and the rest of the town.
Punggol is able to meet the minimum planning standards (Table 4-6) for an LRT system
within a new town.
Table 4-6
Planning Guidelines for Light Rail Transit (LRT) versus Actual
Provisions in Punggol
Parameters
Planning Standards
Actual Provisions in Punggol
Mihimum Population
Population Catchment radius
Station Spacing
Minimum no. of dwelling units
97,500 persons
300 to 350m from proposed
stations
400 to 600m
30,000 (high density)
237,250 persons
300 to 350m from proposed
stations
400 to 600m
73,000 (high density)
FAR
Net housing yield
2.8 minimum
160 du/ha minimum
2.1 to 3.5
197 du/ha
Assumptions
Household size = 3.25
persons/unit 3 1
60% of population (average) make
trips on a daily basis
Average # of trips per day per
person = 1.7
70% of daily trips are public
transportation trips
Population
=
97,500 persons
Average no. of people travelling
on LRT
= 58,500 persons
Average no. of trips per day
99,450 trips
Public transport trips per day
69,615 trips
Population
=
237,250 persons
Average no. of people travelling
on LRT
= 142,350 persons
Average no. of trips per day
241,995 trips
Public transport trips per day
169,396 trips
Source: Urban Redevelopment Authority (undated)
Density determines the walking distance for homes around the stations (Figure 4-r10).
The average FAR is 3.4 for the sites around the stations which are part of the two loops in
the network (and south of the semi-expressway). The walking distances to each station
along both loops is 300m. There is even an overlapping of the catchment areas for some
stations, i.e. some homes are within 300m walking distance from two LRT stations. For
stations along the spur line, the proposed FAR of 2.1 to 2.8 is lower than that of the loop
stations along the spur line, the proposed FAR of 2.1 to 2.8 is lower than that of the loop
lines. The walking distances to the stations along the spur line is 350m, further than that
of the loop lines. This suggests that the higher the residential density around the station,
the shorter the proposed walking distance between homes and the stations will be.
Figure 4-10
MRT & LRT Network and Density
Apart from the policy to provide an LRT system within the town, the DGP is silent on the
treatment of other transportation facilities such as roads and car parks. With the
provision of the LRT, it is interesting to note that there has been no reduction in the width
of road reserves. This suggests that roads still occupy as much space as they would
without an LRT system although the LRT system running on elevated tracks replaces the
feeder buses which run on roads. Does this suggest an optimal use of land? This
question will be explored more fully in the next chapter.
Parks and Gardens
Policy 5 - Provide a network of green connectors and promenades along the rivers
Green connectors (Figure 4-11) are
Figure 4-11
proposed to link the parks, nature sites
Green Connector along Sungei Api
Api in Pasir Ris New Town
and waterfront with residential areas.
They serve as recreational facilities (i.e.
cycling, jogging), as well become
convenient routes for commuting.
Providing a system of connectors does
not require additional land to be taken up
because they utilize existing drainage,
foreshore and road reserves.
Figure 4-12 shows the distribution of the
parks, gardens and open spaces for both
DGPs. The pattern emerging in the case
of Punggol vastly differs from that of
Pasir Ris. Punggol's network of green
connectors is very developed.
Source: UrbanRedevelopment Authority (1995)
94
Each common green and open space is linked to via the system of connectors, which
leads to the green promenades along the waterfront. Although Pasir Ris has an already
well-developed park connector system along the rivers Sungei Tampines and Sungei Api
Api, the relationship between the parks and connectors is not evident. From the green
connector system proposed for Punggol 21, it is evident that the objective of this policy
has been achieved.
After studying the connector network, it is also useful to take a look at the distribution of
parks, gardens and open spaces within the 2 DGPs. The reorganization of the housing
neighborhood from 6,000 dwelling units to 1400-2800 dwelling units in Punggol 21 has
resulted in the parks being reduced in size from the previous standard of 1.4 hectares to
0.4-0.7 hectares. These smaller parks are distributed more regularly throughout the
whole town, allowing residents better access to them because they are within close
proximity of homes.
Table 4-7 shows the land area proposed for parks and gardens within Punggol and Pasir
Ris DGPs. In terms of land consumption, the area dedicated to parks is almost the same
for both Punggol and Pasir Ris. Though park space per capita is lower in Punggol (0.32
ha/1000 persons3 2 vis-a vis 0.56/1000 in Pasir Ris), this is expected because the
population size is much larger (the larger size of the regional park in Pasir Ris also skews
figures).
The overall park strategy for Punggol DGP with smaller but better distributed parks and
open spaces that are linked up through a network of green connectors has good potential
to positively impact the quality of life of the residents, and at the same time consumes
less land.
Figure 4-12
Park and Open Space Network with Green
Connectors - Punggol and Pasir Ris DGPs
Punggol
Pasir Ris
4.
.1
-
*
*1'1'.~ -
p
O
'
2010 00
W
500MEE
0 ~l
-. 1.
Table 4-7
Park and Garden Provision - Punggol and Pasir Ris DGPs
Typology
REGIONAL
Regional Park
Provision
Standard
(DGP)
(ha/1000
persons)
Site Area
Standard
(ha)
Punggol
Area
(ha)
0.74
>30
Coney
Island
Percentage
(%)
33___
Town Park
Percentage
(%)
Pasir Ris
59
150
DGP
Pasir Ris
Area
(ha)
Park
179
78
34____
0.08
10
22
22
13
13
0.02
3
-
-
-
-
0.12
1.2-1.5
-
-
9
9
Common Green
Not
0.4-0.7
13
16
-
-
(1200-2800 du)
(PUNGGOL)
available
TOTAL
-
100
101
100
(40,000 du)
Town Center
Garden35
(40,000 du)
Neighborhood
Park (6000 du)
(PASIR RIS)
-36
Source: UrbanRedevelopment Authority (1992, 1995 & 1998)
Next Chapter
In the course of analyzing Punggol's land use policies and how they contribute towards
optimizing land use, some questions have been raised. The next chapter seeks to explore
these issues in greater detail, as well as propose alternatives strategies to optimize land
use.
1Some of the information contained in the DGP has been superseded by the Master Plan 1998, gazetted on
24 December 1998. Nevertheless this plan will be used because it reflects the general planning intention
for the Punggol area.
2 Sungei in the Malay language means "river".
3 The third "river" will be created between Coney Island and the mainland through reclamation of the
harbor.
4 Urban Redevelopment Authority and Housing and Development Board, 1996 (Punggol 21)
5The Executive Condominium scheme was conceived to allow more Singaporeans to achieve their
aspiration of owing private property. 86% of Singaporeans live in public housing (HDB, 1998). E-condos
are condominium housing developed by HDB, but earmarked for privatization 10 years after the date of
completion. The sites are developed with facilities similar to those in private condominium developments,
such as swimming pools, tennis courts, clubhouses etc.
6 Condominium status is granted to a development only if its plot size exceeds 0.4 hectares (4000 square
meters). If the plot size is less than this figure, the development will be considered a flat development.
7 Town clubs, sports clubs, beach clubs and sailing clubs
8 URA completed the preparation of the Pasir Ris DGP in 1995. As in the case of Punggol DGP, this plan
may have been superseded by the Master Plan 1998.
9 Pasir Ris had 32,988 housing units in the 1990 when the last population census was carried out.
10 99.3% of the public housing apartments in Pasir Ris are 4-room flats or larger. 4-room, 5-room and
executive flats comprise 38.9%, 32,9% and 27.5% of the housing stock respectively. The respective unit
sizes (average) are 100 m2, 125 m2 and 140 m2.
"1This figure excludes the land areas of Subzones 1 (Pasir Ris West) and 6 (Loyang) of the Pasir Ris
planning area.
12 Ministry of National Development
13 ibid.
" Urban Redevelopment Authority, 1998
is Urban Redevelopment Authority, 1995
16 Urban Redevelopment Authority, 1998
" ibid.
18 Pasir Ris does not have an LRT system within the town. There is one MRT station.
19 Derived from Net housing yield x Occupancy Rate
20 Landed housing = 1 dwelling unit per plot
of land
21 Non-landed housing includes strata landed housing (strata bungalows, townhouses
and cluster houses),
residential flats and condominiums. More than 1 dwelling unit is proposed per plot of land.
22 For example, the minimum plot size for the detached house was 550 m 2 as compared
to today's
minimum plot size of 400 M2 .
3 Proposed - Control
2 Landed housing = 1 dwelling unit per plot of land
25 Non-landed housing includes strata landed housing (strata bungalows, townhouses
and cluster houses),
residential flats and condominiums. More than 1 dwelling unit is proposed per plot of land.
26 Total land area designated for residential use is 530ha. However 145 ha is vacant
and subject to detailed
planning. Hence the area is not included.
2 Standard flats are designed in-house by HDB.
28 Design Plus flats are considered premium public housing. They are also designed
by HDB. The
difference between these flats and standard flats are in terms of size and cost. The units are slightly larger
in size than standard flats and cost 10-15% more. They also come with finishes such as flooring (instead of
just concrete) and floor to ceiling windows.
29 Design Plus flats are considered premium public housing. They are also designed by HDB. The
difference between these flats and standard flats are in terms of size and cost. The units are slightly larger
in size than standard flats and cost 10-15% more. They also come with finishes such as flooring (instead of
just concrete) and floor to ceiling windows.
30 According to the Land Transport Authority, the proposed LRT lines for the two new towns of Punggol
and Sengkang, comprising 24 km in length of tracks and 33 stations in total, will be built at a cost of
Singapore $656 million (US $386 million).
31 This is the projected figure for the Year 2000 in the
Concept Plan.
32 Two other reasons make Punggol's park provision appear lower than it is.
Punggol has open spaces
which are proposed in lieu of parks within residential neighborhoods. As these areas are not included as
park areas, the provision standard is artificially lower. The Pasir Ris regional park is 29 hectares larger
than the regional park in Punggol. This also skews the figure upward for Pasir Ris.
33 The town park serves the entire DGP. Where there is a regional park within the DGP, the town park has
to be well located to prevent duplication of functions
Punggol does not have a formal Town Park on a regular shaped plot. Instead the park for the town is in
the form of a linear coastal park. A 15.Om wide promenade is proposed along the 3 rivers (21 hectares
altogether), and a riverine park (1 hectare) is added to create variety.
3 This park functions as the formal garden at the town center as well as a neighborhood park for the area.
36 The land zoned open space is not included in this figure. However the open space appears to substitute
for parks, which are landscaped areas. With the inclusion of these open spaces, the provision standard per
person increases.
3
5
Strategies to Optimize the Use of
Land
This chapter seeks to explore the questions arising from the discussion in the previous
two chapters that are important to optimization of land use, but which the policies have
not addressed. It will look in detail at two issues - the classification of housing variety,
and the pattern of land consumption. In relation to land consumption, additional
strategies to optimize land use will be proposed and evaluated to assess their feasibility.
5.1
Classification of Housing Variety
The policy to introduce a greater variety of housing types will be revisited in this section.
As highlighted in the previous chapter, the policy statement does not clearly articulate
what constitutes variety. From the case study, variety was defined in two ways: by land
tenure and by design quality. However these two categories do not fully capture the
diversity of housing types offered in Punggol 21 and elsewhere.
A variety matrix is proposed to classify housing variety (Figure 5-1). Variety can be
categorized by 9 classes, namely by tenure, property title, property type, site area and/or
unit size, density by residential typology, density by FAR, story height, design quality
and location. Each category and its relationship to housing distribution is more fully
described in Table 5-1. The matrix shows that the first 8 categories overlap with one
another to some degree. Location is the one category for which no clear theoretical
relationship can be established with other categories.
101
Figure
5-15-1 Variety Matrix for Housing
Figur
Ir.l.)
A~l~ j-A~
ILJ
Lj
I~
e
"HOUS I Nj
I t-ni -rLE
LARMD
I
CLANE:)
1
itcom
D-15
0UIY%m,4SITY
F-LoOR. APE
NON C3IATA-TITLE-)
(5
i
c~io~ui
~0
~jo.4h&
*00
Jow
PZA-TIO
PL~la
AtE
A~e
<o4~o.
______
I___
jZZZ
L
coT4
NO~~C
1-6
t __
-
2-1
_
H E I'HT(STCW~S)
2-'
PESI GrEV
LOCA'tXON
TEA3
IVL.vI MWAS1I~Q STAWDAS
t40 CLEAV& REL"fOMSHV'
T
Table 5-1
Classification of the Variety of Housing Types
Category
Distribution
Remarks
SOCIO-ECONOMIC
Tenure
Property Title
Private housing
Public housing
Landed housing
Land Title
Non-landed housing
= Strata Title
Strata Lease
Property Type
Detached house
Site Area
(minimum) and/or
Unit Size
Semi-detached house
Terrace house
Cluster house
Townhouse
Shophousel
Condominium
Flat
Detached house
2
Good Class Bungalow Area
2
(GCBA) = 1,400 m 2
;
Non GCBA = 400 m
2
Semi-detached house = 200 m
Terrace house = 200 m2 (comer) /
150 m2 (intermediate)
Cluster bungalow = 400 m2
2
Townhouse = 150 m
Condominium = 0.4 hectares
Flat < 0.4 hectares
Housing mix was expressed in these terms at
national level.
The availability of on-site recreation facilities is
directly related to property title. Landed housing
and strata lease properties have no on-site recreation
facilities. Strata title properties usually have on-site
communal recreation facilities on-site, such as
swimming pools, tennis courts, clubhouses etc.
Site area is directly related to property type (see
Variety Matrix in Figure 4-4).
Public housing is differentiated by unit size
(number of rooms).
DENSITY
Residential
Low Density
Typology
Medium Density
FAR
: <= 1.4 FAR
Low
Medium: 1.6 - 2.1 FAR
:>= 2.8 FAR
High
See Figure 2-3a for mix of housing units by density
in FAR for Punggol DGP.
: <= 4 stories
Low
Medium: 10-20 stories
: >= 20 stories
High
Height is directly related to residential density in an
attempt to ensure that prescribed FARs can be
achieved.
For public housing units only.
Categories are:
=Standard flats 3
=Design plus flats 4
-Design and build flats 5
In Punggol each type has a one-third share in the
public housing component.
Proximity to:
Waterfront
Parks & Gardens
Recreation areas
*
Cultural / Entertainment areas
Mixed Use
Small-Office-Home-Office
Variety was determined in these terms in the 1991
Concept Plan overview.
High Density
HEIGHT
Story Height of
Building
DESIGN
Design Quality
LOCATION
Location
In the Punggol DGP, variety by location can be
considered in terms of waterfront location, housingby-the-park, and proximity to recreation areas.
103
5.1.1
Housing Variety by Location
Location will be explored in greater detail using the case of Punggol 21. Punggol's
situation is fortuitous because there are physical assets such riverfronts, coastal
waterfronts and a large regional park at Coney Island. In the 1991 Concept Plan
overview, one of the aims was to provide better locations for housing - along waterfronts,
and close to parks, gardens, recreational areas, cultural and entertainment areas. Figure
5-2 shows the proportion of residential land within a 200m radius of the waterfront and
the proposed recreation areas in Punggol. When matched against land tenure, there is a
high correlation between "choice" location and private housing. This is hardly surprising
since good location generates a higher land value, and is therefore assigned a higher-end
residential use. However this raises the issue of equity because land parcels designated
for private housing are located closest to the waterfront. One way to mitigate this is to
restrict the heights of buildings closest to the water's edge to low or medium-rise height
ranges. This allows the taller buildings further away from the waterfront to enjoy views
of the water on the higher floors.
Figure 5-2
Housing Variety by Location and Land Tenure
Public Housing
. .:
Q
Private Housing
200m from
Waterfront
200m from Recreation
Facility
10
104
300
500METER
5.2
Land Consumption
The impact of land use policies on land consumption is an important consideration in
land scarce Singapore. This section will look at the general patterns of land consumption,
and explore in greater detail how housing variety and transportation impact land
consumption. Alternative strategies to optimize the use of land will be proposed and
evaluated.
5.2.1
GeneralPattern
The general pattern of land consumption was briefly discussed in Chapter 3. The trends
observed will be examined more fully here. It is important to bear in mind the
significance of the increase in living standards and people's expectations for a better
quality of life on land take.
Trend 1 - PlanningStandards and Land Consumption
Land take is directly related to changes in planning standards, which include provision,
site area and spacing standards (discussion in Chapter 2 refers). Site area and spacing
standards will first be discussed. Over time, these standards have been relaxed because
of land constraints. The minimum plot areas have been reduced for different building
types (i.e. landed housing). Spacing standards have also been revised to allow buildings
to be built with less set back from plot boundaries. The relaxation of these standards
points towards the reduction in land take. A certain level of environmental quality has to
be sacrificed in order to obtain a higher level of performance from each plot of land. This
trade off has to be considered when these standards are set.
Provision standards at the scale of the city or new town have a significant impact on land
consumption. Over time, people expect provision standards to have become more
stringent because they want a higher quality of life. For instance, the number of
performance arts theatres per capita would increase. With this increase in the number of
facilities or land area (numerator), land take can be expected to increase. The change in
105
the numerator is usually the conscious result of policy decisions towards improving the
quality of life. This is a result of the shrinking of the figure in the denominator
(population) while the numerator stays the same (facilities, land area). For example, the
park space per capita can increase over time if the population size shrinks though no new
land for parks is added (see trend 1). Hence it is important to differentiate which of these
two factors cause provision standards to rise. When setting provision standards, the
challenge facing planners is to find the balance between land consumption and the
desired quality of life level.
Trend 2 - PopulationDensity and Urban Density
Between 1970 to Year X, Singapore's population density has been rising while the urban
density has been falling (Table 3-3 refers). Between 1991 and the Year X, the total
population is projected to increase by 58%, while land dedicated to living space has
increased by 108% (Table 3-4 refers). This points to the fact that land take is growing at
a faster rate than population growth. It can be inferred the higher living standards and
people's expectations for a better quality of life have contributed to this pattern of urban
growth. The boundary of the urban area will continue to expand over time (see Figure 22 showing the urbanized areas in the 1971 and 1991 Concept Plans). Figure 5-3 shows
the growth of the urbanized area between 1970 and Year X.
This trends illustrates that improving the quality of life and reducing land take seem to be
on opposing ends of a continuum. Conventional wisdom suggests that in order to
improve the quality of life, more land needs to be consumed. This implies that the urban
edge needs to keep expanding. Before considering if land scarce Singapore afford this
rate of urban growth, it is appropriate to look at the second trend, the growth pattern of
residential density, which examines how efficiently land in urbanized areas is used.
106
Figure 5-3
Level of Urbanization (1970 to Year X)
Level of Urbanization
80%
70%
60%
50%
40%
30%
20%10%-
0%-1970
1990
1999
2010
2030
Year
Source: Urban Redevelopment Authority (1991)
Trend 3 - Residential Density of New Towns and Global Density
Punggol 21 has an average residential density of 562 persons per hectare (ppha). The
density is more than two times higher than that of Pasir Ris, which is 272 ppha. As a
typical new town of the 2l5' century, the density pattern of Punggol points towards the
trend of developing new towns at residential densities that are much higher than existing.
It is important to note that this trend of development goes against the overall growth
pattern of residential density for the whole country, which has been decreasing over time
(Table 3-4 refers). The overall residential density in 1991 was 418 ppha, and is projected
to fall to 318 ppha in the Year X.
This pattern of residential density suggests that newly developed areas have been
proposed with higher densities to offset the lower densities in the urbanized areas of the
island. As new towns contribute to the spread of the urban edge, the question which
arises is whether it is necessary to keep building more new towns for housing given that
107
the overall residential density is falling. Another issue to consider is if developing new
towns at higher densities adversely affects the quality of life of their residents.
-
Urban Edge
Is there scope for existing urbanized areas to be developed more intensively in order to
contain the spread of the urban edge? This will preserve the remaining vestiges of rural
areas on the main island and the offshore islands. These natural assets will contribute to
enhancing the quality of life.
The 1991 Concept Plan proposal to expand the urban edge has been challenged by
Singapore architect, Tay Kheng Soon6 . He contended that the additional population to
make up the 4.77 million for the Year X could be accommodated in the central region7 .
New housing can be added in the existing private and public housing estates, and by
building dense clusters on undeveloped parcels of land in already urbanized areas. This
position has similarly been echoed by Edmund Waller8 of the National University of
Singapore, who observed that Singapore could have saved more land in the interior by
building new towns along the coast and on reclaimed land. He made comparisons
between Singapore and Hong Kong, two territories with similar development pressures
and land constraints. While Hong Kong has 40% of land area protected as national parks,
Singapore has only about 5% in comparison. Hong Kong's interior has been left
untouched owing to the topography, and development has been concentrated along
coastal inlets and on reclaimed land. This yields the same effect as concentrating
development within urbanized areas. These criticisms of the Concept Plan proposal to
keep expanding the urbanized area are valid as this pattern of land consumption has
implications on the quality of life. The pattern of urban growth has to be carefully
considered by Singapore's planners in the preparation of the next Concept Plan.
-
New towns
Punggol 21 represents an attempt to improve the quality of life while reducing land
consumption. This is evident in the policies to cluster community facilities and
reorganize the layout of residential neighborhoods and the sizes of parks and gardens.
However this attempt to reduce land consumption operates within the framework of the
108
1991 Concept Plan, which projects the spread of the urbanized area. Hence the impact of
these measures on reducing land take is less significant as compared to policies which are
designed to address the underlying issue, the spread of urban edge.
5.2.2
Housing Variety and Land Consumption
Planning policies at the macro level aim to have a more varied housing mix. In view of
this, it would be apt to explore the relationship between housing variety and land
consumption. Since housing types are generally differentiated by density, the
relationship between density and land consumption at the national level, and in the
Punggol and Pasir Ris DGPs will be studied.
Figure 5-4a illustrates the proportion of housing type proposed for the Year 2010 and
Figure 5-4b shows the land area consumed by each housing type. In both DGPs, highdensity housing consumes the most land, followed by medium and then low-density
housing.
This pattern of land consumption does not correspond with the citywide
pattern where high-density housing consumes the most land, followed by low-density
then medium-density housing. However the pattern is typical of any new town since
low-density housing units comprise a small proportion of the proposed housing stock as
compared to areas where landed housing is very established.
High-density housing constitutes a major proportion of the housing stock in both DGPs.
Although its share of land take is lower in proportion than its share of the housing stock.
Medium density housing contributes about the same share to housing stock as it
consumes land. Low-density housing contributes the lowest share of housing stock but
the proportion of land it consumes is the highest. The conclusion that can be drawn is
having a more varied housing mix by density is land intensive. Thus planners have to
carefully consider the housing mix in view of the impact on land consumption. One
option would be to try and increase variety in less land intensive ways, such as by
location.
109
Figure 5-4
a. Distribution of Housing Type (Year 2010)
b. Housing Type by Land Area
Housing Type by Land Area
Distribution of Housing Type
MeiunDeity
* Ndicn~d
EPuW~
*Punggpl
OP~irRis
25%
Low DErity
* Ncticnd
Hgh D
ity
1]3Pair Ris
23%
Low Damity
MedirnDrity
Deity
ghDersity
Dersity
Source: Urban Redevelopment Authority (1991, 1995, 1998)
5.2.3
Transportation and Land Consumption
Roads and infrastructure are the second highest consumer of land in Punggol, taking up
15% of the total land area. As mentioned in the previous chapter, the LRT system has
not resulted in the change in road standards (roads and car parks). Can these standards be
reduced in view of the comprehensive public transportation infrastructure being proposed
for the town? In this section, car parking and road spacing standards will be discussed to
see if changes in standards can be wrought to reduce land consumption.
110
Car Parking Standards
The minimum parking requirement for residential land is one space per dwelling unit for
both public and private housing developments (Table 5-2). With 85,800 dwelling units
proposed for Punggol, 85,800 parking spaces need to be provided on residential land.
Most of the public housing planned for the town will be 4-room (HDB typology) or
larger, and car parks will occupy a significant proportion of the residential land.
Table 5-2
Planning Standards for Public Housing and Land Required for
Transportation Facilities
Flat type
Density
Car Park
Percentage of land needed for
1-room
2-room
3-room
4-room
5-room
Combined
Du/ha
300
235
185
160
160
208
Ratio'
5.5
2.1
1.5
1.0
1.0
1.97
Car Parks
17.0
26.4
31.7
36.2
35.2
30.2
Roads
9.1
9.1
9.1
9.1
9.1
9.1
Total
26.1
35.5
40.8
45.3
44.3
39.3
Source: Olszewski, 1983
Figure 5-5 shows a typical layout for HDB flats in Punggol 21 (FAR 3.4). Multi-story
car parks are proposed to meet parking requirements, and these structures are 4-stories
high. They will be integrated with the apartment blocks. In Singapore, car parking
structures are exempted from Gross Floor Area (GFA) computation. This means that the
3.4 FAR is for the GFA of the residential blocks only. If the parking structure is included
in GFA, the FAR will exceed 3.4. From the layout, it is noted that the coverage of the
site is very high. The multi-story parking structure itself sits on 15% the ground. The
environment is very built-up. To mitigate the intensity of development, designs are
proposed where the roof of the parking structure will be landscaped and used as a garden
and recreation area. The roof is also part of the pedestrian network that links all the
blocks within the same neighborhood, improving the pedestrian-friendliness of the
environment.
111
Figure 5-5
Typical Layout of Public Housing in Punggol 21
Notes
MSCP = Multi-Story Car Parking
Numbers on the plan, i.e. 101, 105 etc are Apartment Block Numbers
Source: Urban Redevelopment Authority
As car parking facilities take up such a high proportion of residential land, a reduction in
the parking standards will contribute significantly to the reduction in the consumption of
land. With the proposed public transportation infrastructure (MRT and LRT) promising
seamless travel within the town and beyond, the question that comes to mind is whether a
good proportion (i.e. 20%) of Punggol 21 households will forego ownership of the
private car because public transportation is a good substitute for travel.
112
The good public transit system may encourage residents to reduce the number of trips
made by car in lieu of transit. Nevertheless, parking spaces on residential land still have
to be provided if a household owns a car, regardless of the number of trips made with the
car. Reducing parking standards will not be viable unless public transportation becomes
a full substitute for the car for a proportion of the households. With the rising affluence
of the population and the aspirations for a better quality of life, this may be an unrealistic
expectation. Car ownership patterns have to be carefully studied as parking standards are
tied to them. Unfortunately this is not within the scope of this study.
Road standards
It is apt to turn to road standards to see if they can be revised to bring about land savings.
In view that travel people's patterns may be altered in favor of more public transportation
trips, reducing the width of travel lanes and/or road reserves may prove to be a viable
option to reduce land take. To set the stage for discussion, some general transportation
indicators for Punggol and Pasir Ris DGPs are set out in Table 5-3. The road network for
the two DGPs is shown in Figure 5-6 and Figure 5-7 shows the road types for Punggol.
113
Table 5-3
Roads - Punggol and Pasir Ris DGPs
Road System
Road Type
Road Reserve
(Width in
meters)
Punggol DGP
Road Length
(meters)
Pasir Ris DGP
Road Length 0
(meters)
Expressway
Expressway
52.9
) 3,500 (7.5%)
) 4,700 (10%)
)
)
) 3,500 (7.5%)
) 1,050 (2%)
)
) 900
(1.8%)
System
Arterial System
Semi-
45.5
45.4
Expressway
38.6
Major Arterial
45.4
)
) 1,500 (3%)
38.6
)
)
38.6
) 12,020 (26%)
) 11,850 (24.5%)
31.8
)
)
31.8
) 14,930 (32%)
) 11,840 (24.5%)
26.2
21.4
)
)
)
)
18.0
15.4
) 11,450 (24%)
)
) 18,020 (37.2%)
)
14.2
12.2
11.6
11.1
)
)
)
)
)
)
)
)
46,900 (100%)
0.18
48,360 (100%)
0.33
Land Area
(hectares)
Land Area
(hectares)
Arterial
Collector System
Local Road
System
Primary Access
Local Access
TOTAL
Per Capita
-
-
-
-
Roads
All Types
-
131
-
0.49 m2
-
Housing Stock
85,800 du
46,400 du
Occupancy Rate
Total Population
3.11 ppu
266,838 persons
3.11 ppu
144,304 persons
TOTAL
Per Capita
-
Background Information
114
Figure 5-6
Road Network - Punggol and Pasir Ris DGPs
Punggol
Pasir Ris
Not to Scale
115
Figure 5-7
Road Types - Punggol DGP
Expressway
Semi-Expressway
-irr
Major Arterial
~tn
Arterial
-
Primary Access
Local Access
i
~~
0
40
200
10
0
00METER
Reducing the standards for access roads within the collector system and local road system
is a possible a strategy to save land. These roads, whose main function is to provide
access, are good candidates for consideration because road capacity is a secondary
function. They can tolerate slower travel speeds, and hence there is potential to reduce
the width of the travel lanes. Alternatively we can consider reducing other components
of the road reserve such as the planting verge, service verge and footway. The road
reserve for access roads comprises the carriageway for travel, which is in the center. On
both sides of the carriageway, there is a verge for services and tree planting, and a
footpath (over the drain). Figure 5-8 shows a cross-section of a typical road reserve.
116
Figure 5-8
Typical Road Reserve (Section)
LoT
b I FFEt
I-to A
-tgI
wLMAt.Y
-LOT BoUNPA5
I
. -I
-,FP~
-4ZAV
3wq-IO.6w~
0
Source: Urban Redevelopment Authority (1992)
-t4rA
30-6r
I-
ivoy
IOBM-AsmI3.5w~
so M1
.
-a-t~ m
-4--- IOM
-4
This analysis involving the reduction of one or more components of the road reserve will
use the Punggol DGP as the case study. Although both the road length per capita and the
per capita area occupied by roads is lower in Punggol than Pasir Ris DGP, there is still
potential to consider reducing road reserve widths in order to assess their impact on land
savings for the DGP. The proposal involves roads in the collector and local road
systems. In the Punggol DGP, they constitute a significant 56% of the total road length.
Five options are proposed. For each option, one or more components of the road reserve
will be changed. The options are explained below, and a summary with illustrative
figures showing the proposed changes is given in Table 5-4.
= Option 1
This first option proposes to completely remove the service and planting verge within the
road reserve. The service verge can be dispensed with as essential services have been
placed underground. The exception is the street lamps, which can still be accommodated
with the reserve. The planting verge can also be removed. This is because of the
planning requirement that a buffer has to be provided between a building and the road.
The required green buffer on private land can be a substitute for the planting verge.
-
Option 2
The proposal involves combining the service and planting verge and reducing the width
from the present standard. This ensures that there will be planting along public roads, in
keeping with the Garden City image Singapore wants to preserve. In addition, it takes
into account that road reserves form part of the proposed islandwide park connector
system linking parks and natural areas to residential areas. The area of these connectors
contributes to satisfying the overall park provision standards, which is 0.8 hectares/1000
persons".
118
-
Option 3
This proposal involves the reduction in the widths of travel lanes. As road capacity is
secondary to access, narrow travel lanes and slower travel speeds are acceptable. The
slower speeds are welcome in residential areas as they encourage safety. The width
reduction depends on the type of road, and the absolute minimum width is maintained at
3.0m.
-
Option 4
This option is the most aggressive of the options proposed as it is a combination of
options 1 and 3, resulting in the highest land savings. The lane width is reduced, and
service and planting verges completely eliminated.
-
Option 5
The final option combines options 2 and 3. Both travel lanes and planting and service
verges have smaller widths than existing.
Analysis
The land savings for each option are shown in Table 5-5. The lower limit of land saved
is given in the column "LRT" for each option. Where the LRT track runs, the planting
and service verges are reduced to one per road reserve instead of being completely
removed for the purpose of reducing the disamenity of the train line running close to
residential buildings. Hence this figure can be considered the lower limit of land savings,
while the "no LRT figure" is the upper limit. The amount of land saved will fall within
this range, depending on the length of road with LRT lines.
Option 4 yields the highest savings in land, followed by options 1, 5, 2 and 3
respectively. Although it is attractive to completely do away with the planting verge as
proposed in options 4 and 1, relying on the green buffer on development sites for planting
along roads may be problematic, especially for roads abutting privately owned land.
Firstly, boundary walls are usually erected along the road reserve line. This screens the
119
green buffer from public view. As planning permission is not required for the building of
a boundary wall, it may be difficult to replace the planting verge along public roads with
private planting verges. This is not an issue along roads adjacent to public housing plots
because no boundary walls are proposed. Another problem will be maintaining the
garden city image as individual landowners are charged with the task to maintain the
greenery.
The recommendation is to consider option 5, where travel lane widths are reduced and
the planting and service verges combined into one. The narrower planting verge can be
more lushly landscaped to preserve the green city image, and environmental quality need
not be sacrificed. Another option is to switch the alignment of the roadside table such
that the footway and drain is adjacent to the carriageway instead of the planting verge.
The reduced planting verge can then be combined with the green buffer on development
sites to become a bigger area for planting. However the implications have to be studied
in order to ensure that having the footway next to the carriageway does not compromise
pedestrian safety and health. This proposal warrants further study in order to fully assess
its viability.
120
Table 5-4
Proposals to Reduce the Width of Road Reserve
Option
Road Reserve Components (Widths)
Illustrative Example
1
Lane
No change
Service Verge
Remove
completely
Rationale:
Planting Verge
Remove
completely
Footway
No change
2
No change
Combine service and planting verge
No change
3
Reduce to 3.03.3m
No change
No change
No change
4
Reduce;
Dimensions
Remove
completely;
Remove
completely
No change
the same as in
Option 3
5
Reduce;
Dimensions
ROAD PESEQVE 31 8m
yc
F
P
.
Combine service and planting verge as
in Option 3
No change
-t
[.0
St?
P
550
WITH
C
t
the same as in
Option 3
psopox
t
n
-
SCALE :: ooA
Table 5-5
Reducing the Width of the Road Reserve and the Resultant Land Savings
ROAD TYPE
Primary Access
A
31.8
26.8
21.8
29.8
27.8
30.2
30.2
25.2
20.2
28.2
26.2
B
C
26.2
21.4
22
18.9
17.8
16.4
24.2
21.4
22.2
18.9
25.2
20
25.2
20
21.0
17.5
16.8
15.0
23.2
20
21.2
17.5
18.0
15.5
13.0
15.0
14.0
18
15.5
15.5
13.0
15
14
B
15.4
-
14.2
-
-
12.4
11.8
-
14
12.8
D (Terrace houses -
12.2
-
9.0
-
10.2
-
11.6
-
9.0
8.4
8.4
-
11
C
10.4
9.8
10.4
9.6
11.6
-
8.4
-
9.6
-
11.6
-
8.4
-
9.6
11.1
-
7.9
-
9.1
-
11.1
-
7.9
-
9.1
-
476,631
401,836
46,614
85,544
38,151
22,326
110,473
185,268
69,536
108,466
-
7%
12%
4%
7%
3%
2%
8%
14%
5%
8%
Access
Local
A
(Public housing)
frontage)
E (Semi-detached &
detached houses frontage)
F (Semi-detached and
detached houses - cul-
de-sac)
LAND SAVED
(M2)79
Savings as a
percentage of total
area occupied by
roads
9
A shophouse is terrace building (common party walls along the side plot boundaries) with commercial use
on the ground floor and residential use on the upper floors. Access to the upper floors may be segregated
from the access to the ground floor, or may be through the ground floor units. This type of housing is the
traditional form found in the historic districts and other local shopping areas.
2 Good Class Bungalows are detached houses are found within Good Class Bungalow Areas, which are
established landed housing areas in residential districts close to the city center. The minimum site area of
bungalow plots has to be 1,400 m2 (0.35 acres).
3 Standard flats are designed in-house by HDB.
4 Design Plus flats are considered premium public housing. They are also designed by HDB. The
difference between these flats and standard flats are in terms of size and cost. The units are slightly larger
in size than standard flats and cost 10-15% more. They also come with finishes such as flooring (instead of
just concrete) and floor to ceiling windows.
5 Design Plus flats are considered premium public housing. They are also designed by HDB. The
difference between these flats and standard flats are in terms of size and cost. The units are slightly larger
in size than standard flats and cost 10-15% more. They also come with finishes such as flooring (instead of
just concrete) and floor to ceiling windows.
6Straits Times, June 24 1992
7This area is bounded by Marina South, Queenstown, Bukit Timah, Toa Payoh, Geylang and Marine
Parade DGPs.
8 Waller (1998)
9 Car Park Ratio = Number of dwelling units per car parking space.
10Excludes roads outside the study area
" Land Area is counted twice as it is used for roads as well as park connectors.
123
6
Findings and Recommendations
This study began with trying to find an answer to the question if land use planning
policies have been successful in putting land to optimal use. The focus was the study of
measured density, which is used as an indicator of successful urban performance.
However density cannot be viewed as an isolated variable, but has to be considered in
relation to planning standards and land take in view of the close inter-relationship
between these constructs.
The aims of the study were to assess the suitability of density measurement methods used
in Singapore, and how successful the current practice has been in optimizing land use. In
this final chapter, the findings related to this two aims will be discussed, and
recommendations will be made to improve the existing planning system. The questions
which have arisen in the course of this exercise will be highlighted as they represented
potential areas for further study. A summary of the findings, recommendations and
questions arising from the study is given in Table 6-1. Finally the shortcomings of the
study will be addressed.
6.1
6.1.1
Findings and Recommendations
Finding an appropriate set of quantitative density measures for
Singapore
Floor Area Ratio (FAR)
The case study discussed in Chapter 4 showed that FAR is the prevalent method of
density measurement used for all land uses. At the site level, each individual land parcel
is assigned a FAR for a particular use, i.e. commercial, residential, commercial
125
/residential etc. On the Master Plan, density is reflected in the form of FARs. Anyone
reading the Master Plan would come away with the impression that density is measured
in terms of a simple numerical figure. However this clearly is not the case.
Measured density is one of the tools in the planner's kit for preparing land use plans to
guide long-term urban development. Plans are documents which are for the consumption
of planners, as well as the general public. Because they have a public interface, planners
have to consider using a measure of density which is intuitive and clear to users. The
plan needs to speak a language which landowners and the general public can understand.
FAR is a simple way of communicating how intensively land can be developed. Armed
with information on the site area and FAR, any landowner can know how much GFA
they can be allowed to develop on their land. By using FAR instead of population
density (in persons per hectare), URA is able to make long-term planning intentions
transparent to stakeholders. This in turn facilities the control of development as there is
little room for misunderstandings.
However FAR is really the end product of a rigorous process of planning analysis. The
Concept Plan draws up the long term planning intentions and determines the quantum of
housing, commercial space, working space etc for the population. These projections are
then translated at the local planning area level (DGP). In planning a residential new town
like Punggol, planners have to first determine the number of people to be housed there
based on projection of population growth, household size etc. They then decide how
many residential dwelling units need to be planned to accommodate the population, the
quantum of commercial space required to service the population, and the transportation,
infrastructure, education and recreation facilities to be provided for. The provision of
these facilities is determined by the provision and site area standards set by URA. Finally
FARs are assigned to individual land parcels.
Behind the scenes, there is actually more than one method of measuring density. In
planning for residential use, FAR is related to the net housing yield figures. This clearly
126
points to the fact that a system of checks and balances have been built into the system of
measuring density.
The following recommendations to the FAR system to measure density are proposed:
-
Review 1.4 FAR category
Arising from the case study, it is recommended that the 1.4 FAR category for low-density
residential use be reviewed because the proposed number of dwelling units seem to fall
short of the target net housing yield stipulated in the typology for residential use.
However it is important to note that FARs of 1.4 are usually proposed for sites within the
proximity of landed housing areas. The question that arises is how planners can be
sensitive to the amenity and ambience of landed housing estates while trying to optimize
the use of land.
-
Include Net Housing Yield as a measure of Residential Density in Master Plan for
Residential Use, in addition to FAR
In order to establish relationships between different density measures, certain
assumptions have to be made. In the case of residential use, the relationship between
FAR and net housing yield can be determined if the average size of dwelling units is
assumed. As it is important to ensure that the target number of housing units planned for
is met by actual supply, it is recommended that the Master Plan reflect both the
residential density in FAR and in terms of the number of dwelling units per land area.
Global Density Measures
Population and Urban Densities are two global measures of density which are useful
indicators of the how intensively land is used. As mentioned in Chapter 3, urban density
is particularly useful in highlighting the pattern of land consumption for urban
development. This is a good indicator of the size of the land bank, which is important for
a small country like Singapore.
127
Overall, the density measures used in Singapore appear to be suitable in relation to the
development goals.
6.1.2
Planning Practice in Singapore
Pattern of Land Consumption
This trend of the growing urban edge is characteristic of many cities, both in developed
and developing countries. Singapore is no exception although there are more
development constraints than cities in other countries. Between 1970 to the present,
Singapore's urbanized area has expanded from 32% of land area to 58% of land area.
The urban boundary is projected to grow outwards over time to the Year X to occupy
74% of land area. The boundaries of cities in developed countries tend to grow because
there is a demand for a better quality of life whereas those in developing countries grow
because of a high rate of population growth. Singapore faces pressure to cater to a
growing population, as well as improving the quality of life.
Another observable trend is the uneven distribution of residential density within the
urbanized area. New towns like Punggol are much more densely developed than older
new towns such as Pasir Ris, as well as in other parts of other urbanized area. This
suggests the under utilization of residential land within existing urban areas, which again
is characteristic of many cities.
The existing land use planning policies have resulted in these trends. They are alarming,
particularly because the land supply is very finite. The following recommendations to
reduce the consumption of land are proposed for further study:
-
Review the policy to expand urbanized area
The policies to optimize land use examined in the case study operate within the
framework where it is assumed that urban boundary will continue to expand at the
projected rate. It is important to revisit the underlying assumption, although it is
128
unrealistic to prohibit the expansion of the edge. It is recommended that Singapore
consider developing residential land within the existing urbanized areas more intensively
to control the spread of development in the form of new towns to outlying parts of the
island and the offshore islands.
-
Reduce land consumption of land intensive uses
Strategies to reduce land-take should be focused on land uses that constitute a significant
proportion of the land budget, i.e. roads and utilities. Chapter 5 discussed the reduction
of road standards to save land. There is scope to conduct further studies together with the
relevant authorities to assess the feasibility of reducing planting and service verges and
lane-widths within road reserves for access roads (they are part of the collector and local
road systems).
Two specific proposals for study are:
-
Exchanging the position of the planting verge and footways/drains to have the latter
next to the carriageway instead of the planting verge.
-
Reducing widths of travel lanes of access roads. The impacts on travel safety, speeds
and capacity need to be reviewed.
In addition, road standards and other indicators such as car population growth and
parking standards should be benchmarked against those of cities that have innovative
transit-oriented land use policies. Curitiba in Brazil may be a good example to look at, as
road standards have not changed over time in spite of the high rate of growth of the
vehicle population.
-
Review the proposed quantum of low, medium and high-density residential dwelling
units
The policy to provide more housing variety has resulted in the differentiation between
low, medium and high-density housing types, and providing more low and medium
density housing. The housing mix by density needs to be considered in relation to the
129
target population size, and the proportion of low-density housing units will have to be
studied because they are land intensive
Improving the quality of life through housing variety can be achieved in other ways than
having housing units of varying densities. The housing variety matrix in Chapter 5
illustrates this. Planners should consider less land intensive ways to have varied housing
types. Housing variety by location is an example of this.
-
Track changes in provision standards over time
It would be a useful exercise to take stock of the changes in provision standards over time
in view of their impact on land consumption.. Some standards have changed as a result
of policy decisions to improve the quality of life (numerator), while others are the result
of the slower rate of population growth (denominator). Differentiating between the two
causes of the changes would give a clear picture of the impact of standards on land
consumption. The key is for planners to find a set of standards that reflect the optimal
use of land where the level of land take is balanced with a particular quality of life level.
Frequency of use of land
Intensity of use is usually expressed in terms of density. However intensity can also be
considered in terms of the how frequently land is used. The case study showed that the
policy to locate community facilities together and share common facilities recognizes
this. An index to measure frequency of use needs to be developed. This will facilitate
the better programming of complementary land uses such as day and night uses, and
weekday and weekend uses. The costs of a higher frequency of use (in terms of
management and wear and tear, etc.) also need to be evaluated against the benefits
brought about by sharing land. Should the benefits of such a system be greater than the
costs, a review of the existing land use zones and use classes may have to be set in
motion to allow this type of development.
130
Table 6-1
Summary of Findings, Recommendations and Questions arising from the Study
Aim
1
2.
Findings
Recommendations
To find an
appropriate set of
quantitative
density measures
for Singapore
given its unique
development goals
-
-
Review FAR 1.4 band
-
Master Plan / DGPs should reflect residential
density in both FAR and Net Housing Yield
Terms
-
To assess the
success of current
planning practice
in optimizing land
use
a)
-
Review underlying assumption on which
land use strategy (Concept Plan) is based as
this assumption results in the expansion of
the urbanized area to 74% of land area in
Year X. The distribution of residential
density within the existing urbanized area
needs to be evaluated to see if densities can
be increased in certain areas.
Reduce land take of land intensive uses such
as transportation, i.e. roads and car parks.
Review the proposed quantum of low,
medium and high-density residential
dwelling units in relation to the target
population size. Housing variety may be
achievable in less land intensive ways.
Track changes in provision standards over
time to understand their impact on land
consumption.
-
-
-
-
Residential density is measured in
more ways than FAR. Net Housing
Yield (du/ha) is also used to capture
density.
The FAR measure is a simple way of
quantifying density in terms of the
public interface of the plan.
PATTERN OF LAND
CONSUMPTION
The urbanized area has expanded and
will continue to grow outwards over
time.
The distribution of residential density
within the urbanized area is not
consistent. New towns like Punggol
are much more densely developed
than older new towns and other
developed areas.
"
b) FREQUENCY OF USE OF LAND
Land use intensity is considered in terms
of the frequency of use. The intent is to
reduce land consumption.
Develop a quantitative measure of land use
frequency if the benefits of sharing land/facilities
(i.e. how much land is saved) between compatible
land uses outweigh the costs (i.e. managing and
maintaining the shared facilities).
Questions Arising from the Study
-
"
Is urban development at the
pace proposed in the Concept
Plan sustainable?
How does the changing
population profile impact land
consumption?
How does technology impact
land consumption?
Are provision standards flexible
enough to cater to changing
demands of the population?
-
6.2
Questions for FurtherStudy
The questions for further study relate to the land consumption pattern. They are general
questions which should be considered in every city facing the similar trend of the
expanding urban boundary that has been highlighted earlier.
a
Is urban development at the projected rate of growth sustainable?
This is a critical question for any city to ask. Sustainability would have to be defined in
order to explore this question further.
= Population profile and land consumption
Singapore's aging population profile will put new demands on land, and change some
existing ones. What will be the impact on land consumption?
a Population profile and provision standards
The Punggol DGP appears to be planned for a relatively young population. There are a
large number of schools proposed, i.e. 18 primary schools and 12 secondary schools. It is
unclear if planning standards have an in-built flexibility to cater to the changing demands
of the population. This is an important issue to consider because of the implications on
land consumption.
= Technology and land consumption
With the advent of the information age, technology plays a very central role in our daily
lives. A pertinent issue is how technology will affect the demand for land. One possible
scenario is the mushrooming of home offices, changing the demand pattern of land, as
well as unit sizes of homes. This is an important issue in land use planning and warrants
careful consideration.
132
6.3
Shortcomings
The focus of this study is on measured physical density. Nevertheless the issue of the
quality of life has frequently been raised and discussed in the course of this exercise.
Land use policies reflect that this is a key consideration in planning. Quality of life
issues cannot be sufficiently addressed when looking at density from a quantitative point
of view alone.
Punggol 21 is a case in point. Although proposed with residential densities that are
higher than other new towns and park sizes are smaller, new public housing applicants
for 5-room flats have indicated their preference to live in Punggol 21 above any other
new town (the towns are all about equal distance from the city center)2 . The waiting time
for a flat is about 3 years for towns like Jurong West, yet people are prepared to wait an
additional 2-3 years for flats in Punggol. The popularity of Punggol 21 goes against the
belief that people will choose to live in places where the built-up density is lower. This
suggests that having higher measured density may not adversely affect people's
perception of their living environment.
It is evident that the home buyers perceive that Punggol 21 offers them a better quality of
life than elsewhere. The qualitative physical layout of the built environment contributes
significantly in influencing people's perceptions of density. Punggol offers a different
layout of the neighborhoods with the common green as the focal point. At the townscale, each park and open space within the town is connected to one another, to the
waterfront promenades, the regional park as well as to transportation nodes and
commercial centers. This highlights that any discussion on density would be incomplete
if the human element is not taken into consideration. Ultimately, as "users" of built
environments that are planned and created by planners and designers, people's
perceptions of density are very important and have to be understood.
133
6.4
Conclusion
It is evident that much thought has gone behind policies to put land to intensive use in
Singapore. From the case study of the Punggol DGP, planning policies are relatively
successful in achieving the goal of land use optimization as defined within the current
framework of the Concept Plan. However it is important for Singapore to revisit the
underlying assumption that the urbanized area has to grow at the rate projected in the
Concept Plan in order for land to be put to optimal use. In trying to determine the
optimal level of use, a balance needs to be struck between the level of density and quality
of life. In addition to focusing on the quantitative measurement of density, planners need
to consider people's perceptions of density because they are the end users of the built
environment. This will enhance planning practice and bring Singapore a step closer
towards becoming the excellent city that planners aspire her to be.
134
1 Low-density housing consumes almost the same proportion of land as high-density housing but
contributes to only 11% of the total housing stock vis a vis the 76% contributed by high density housing
(Figure 5-5 refers).
2 Straits Times Interactive, May 14 1999
135
Bibliography
Chapter 1
Alexander, E.R and Reed, K.D. (1988), Density Measures and their Relation to Urban
Form, University of Wisconsin-Milwaukee, USA
Liu, T.K. (1991), A City without a Hinterland, Working paper presented at the
International Colloquium of Economic Growth and Environmental Protection, Quality of
Life - "A Challenge on the Borders of Cities", 10-11 September 1991, Chantilly, France
Liu, T.K. (1991), Implementing the Concept Plan, Conference Paper presented at City
Trans Asia 1991 Conference, 17 September 1991, Singapore
Rapoport, A. (1975), "Towards a Redefinition of Density", Environment and Behavior,
Vol. 7 No. 2, pp. 133- 158, United Kingdom
Republic of Singapore (1998), Master Plan Written Statement 1998, Singapore
Singapore Department of Statistics, Singapore Facts and Pictures 1997, Singapore
Straits Times, Feb 10 1999, "Singaporeset to hit 4 million this year"
Urban Redevelopment Authority (1991), Living the Next Lap, Singapore
Urban Redevelopment Authority (1996), URA Corporate Planning Exercise FY97/98;
Project Team A, Unpublished Report, Singapore
Chapter 2
Faludi, A. (1973), Planning Theory, Pergamon Press, London, p.151
Liu, T.K. (1991), A City without a Hinterland, Working paper presented at the
International Colloquium of Economic Growth and Environmental Protection, Quality of
Life - "A Challenge on the Borders of Cities", 10-11 September 1991, Chantilly, France
Liu, T.K. (1991), Implementing the Concept Plan, Conference Paper presented at City
Trans Asia 1991 Conference, 17 September 1991, Singapore
137
Pearson, H.F. (1982), "Lt. Jackson's Plan of Singapore" in Tan Sri Dato Mubin
Sheppard (ed), Singapore 150 Years, Malaysian Branch of the Royal Asiatic Society,
p. 15 1
Public Works Department, Development and Building Control Division (1989),
Development Control Floor Manual of Floor Area Calculations, Unpublished Report,
Singapore.
Republic of Singapore (1998), Master Plan Written Statement 1998, Singapore
Republic of Singapore (1998), The Planning Act (Cap. 232, 1998 Ed) (Act 3 of 1998),
Singapore
Teo, S.E. (1992), "PlanningPrinciplesin pre and post-independenceSingapore", Town
Planning Review, Vol.63 No. 2, pp.163-185, United Kingdom
Waller, Edmund (1998), Towards Singapore's 3rd Concept Plan: Some Landscape
Planning Considerations, Paper presented at the EDITT'98 Seminar, 27 November - 3
December 1998, Singapore
Chapter 3
Alexander, E.R and Reed, K.D. (1988), Density Measures and their Relation to Urban
Form, University of Wisconsin-Milwaukee, USA
Alexander, E.R (1993), "Density Measures: A Review and Analysis", Journal of
Architectural Planning Research, Vol. 10 No. 3, pp.181-197, USA
American Public Health Association, Committee on the Hygiene of Housing (1960),
Planning the Neighborhood, Public Administration Service, Chicago
Cameron, G. (1980), "Housing densitiesfor developing countries:A discussionpaper",
Third World Planning Review, Vol. 2 No. 1, pp.45-52
Cardew, R. (1996), "ResidentialDensities in Sydney", Australian Planner, Vol. 33 No. 2,
pp. 105-113, Australia
Federal Housing Administration (1971, 1965), Land Use Intensity, US Government
Printing Office, Washington DC
Gibberd, F. (1967), Town Design, Architectural Press, London
Hitchcock, J. (1994), A Primer on the Use of Density in Land Use Planning, Papers on
Planning and Design Number 41, University of Toronto, Canada
138
Hoffman, H. (1967), Row Houses and Cluster Houses: An International Survey, Praeger,
New York
James, J.R. (1967), "ResidentialDensities and Housing Layouts", Town and Country
Planning, Vol. 35 No.11, pp. 552-561
Jensen, R. (1966), High Density Living, Praeger, New York
Keeble, L. (1971), "ResidentialDensity and Layouts", Royal Australian Planning
Institute Journal, July, pp. 82-88, Australia
League of Oregon Cities, Urban Service Program and University of Oregon, Bureau of
Governmental Research Service (1977), Residential Density Standards, University or
Oregon, Eugene, Oregon
Loder & Bayly (1994), Urban Density Definitions, in Magri, S. (1994), "Urban Density
Definitions", Urban Futures, Vol.4, pp. 49-53
Martin, L. & March, L. (1969), "Land use and builtforms", Cambridge Research, April,
pp. 8-14
Mcloughlin, J.B. (1991), "Urban Consolidationand Urban Sprawl: A Question of
Density", Urban Policy and Research, Vol.9 No.3, pp. 148-156, Australia
Republic of Singapore (1998), Master Plan Written Statement 1998, Singapore
Republic of Singapore (1998), The Planning Act (Cap. 232, 1998 Ed) (Act 3 of 1998),
Singapore
Stevens, P.H.M. (1960), Densities in Housing Areas, Her Majesty's Stationery Office,
London
Singapore Department of Statistics (1998), Key Annual Indicators, Webpage, Singapore
Singapore Department of Statistics (1998), Selected Historical Data, Webpage, Singapore
Urban Redevelopment Authority (1991), Living the Next Lap, Singapore
Urban Redevelopment Authority (1992), Road Typology, Unpublished Report, Singapore
Urban Redevelopment Authority (undated), Land Use - Prototypes for New Town,
Neighborhood, Estate and Precinct, Unpublished Report, Singapore
139
Chapter 4
Housing and Development Board (1998), 1997/989 Statistics, Webpage, Singapore
Republic of Singapore (1998), Master Plan Written Statement 1998, Singapore
Straits Times, Apr 20 1999, "Travel to be Seamless in Sengkang"
Urban Redevelopment Authority (1992), Open Space Typology, Unpublished, Singapore
Urban Redevelopment Authority (1992), Plot Ratio and Building Height Typologies for
Residential Development, Unpublished Report, Singapore
Urban Redevelopment Authority (1995), Pasir Ris Planning Area Planning Report,
Singapore
Urban Redevelopment Authority & Housing and Development Board (1996), Punggol
21: A Waterfront Town of the 2 1St Century, Singapore
Urban Redevelopment Authority (undated), LRT Planning Guidelines, Unpublished,
Singapore
Urban Redevelopment Authority (1998), Punggol Planning Area Planning Report,
Singapore
Chapter 5
Oleszewski, P. (1983), "Roads and Car Parks in High-Rise Housing Estates - Some
Planningand Design Problems", Planews, Vol.9 No.3, pp.16-22, Singapore
Straits Times, June 24 1992, "No more new towns"
Urban Redevelopment Authority (1991), Living the Next Lap, Singapore
Waller, Edmund (1998), Towards Singapore's 3rd Concept Plan: Some Landscape
Planning Considerations, Paper presented at the EDITT'98 Seminar, 27 November - 3
December 1998, Singapore
Chapter 6
Straits Times Interactive, May 14 1999, "Long Waitfor Flats in Punggol 21"
140